WO2023097749A1

WO2023097749A1 - Flow regulation and control method and system, intelligent terminal, and storage medium

Info

Publication number: WO2023097749A1
Application number: PCT/CN2021/137091
Authority: WO
Inventors: 於陈慧; 郑富德; 刘硕
Original assignee: 同程网络科技股份有限公司
Priority date: 2021-11-30
Filing date: 2021-12-10
Publication date: 2023-06-08
Also published as: CN114255097A

Abstract

The present application relates to the technical field of computers, and in particular to a flow regulation and control method and system, an intelligent terminal, and a storage medium, aiming to solve the problem in the prior art that because the order quantity of different commodity transactions is affected by exposure positions to different degrees and the exposure positions of all stores are regulated and controlled by using a single rule, expected regulation and control effects cannot all be accurately achieved. The technical solution of the present application is a flow regulation and control method, comprising the following steps: obtaining a search request of a user, obtaining an initial commodity list on the basis of the search request, and performing commodity sorting on the initial commodity list on the basis of a preset sorting rule; obtaining a regulation and control demand of a target commodity, and obtaining a regulation and control solution of the target commodity under the search request on the basis of the regulation and control demand; and performing secondary sorting on the initial commodity list on the basis of the regulation and control solution to obtain a regulation and control commodity list. The present application has the effect of improving the accuracy of flow regulation and control.

Description

A traffic control method, system, intelligent terminal and storage medium

technical field

The present application relates to the field of computer technology, and in particular to a traffic control method, system, intelligent terminal and storage medium.

Background technique

E-commerce refers to the high-efficiency trading activities conducted by buyers and sellers online through the Internet in the open network environment of the Internet in the wide range of commercial and trade activities around the world, which can realize online shopping, Online transactions and online electronic payments between merchants and various commercial activities. On this basis, the rapid development of mobile applications has spawned a large number of application platforms for realizing e-commerce services, enabling users to select and trade merchants and their products according to the merchant information placed in mobile applications.

The application platform that provides e-commerce services can help stores obtain a huge source of customers, attracting a large number of stores to settle in, leading to fierce competition for commodities. In a competitive environment, the display order of merchants on the application platform, that is, the exposure position of merchants, has a great impact on the number of clicks and orders of merchants. Stores with a high exposure position can obtain higher user traffic and thus make more orders.

At present, the platform can realize the control of store traffic by adjusting the exposure position. The platform usually adopts fixed rules to control the exposure position of the store, such as top, bottom and adjustment to a specified position.

In the process of implementing the present application, the inventor found that the above-mentioned technology has at least the following problems:

The order volume of different commodity transactions is affected by the exposure position in different procedures. Using a single rule to regulate the exposure position of all stores cannot accurately achieve the expected regulatory effect.

Contents of the invention

In order to improve the accuracy of traffic regulation, the present application provides a traffic regulation method, system, intelligent terminal and storage medium.

In the first aspect, a traffic control method provided by the present application adopts the following technical solution:

A flow control method, said method comprising the following steps:

Obtain a user's search request, and obtain an initial product list based on the search request, and the initial product list is based on a preset sorting rule for product sorting;

Acquiring the regulation requirements of the target commodity, and obtaining the regulation plan of the target commodity under the search request based on the regulation requirements;

Based on the control scheme, the initial product list is sorted twice to obtain the list of regulated products.

By adopting the above technical solution, after obtaining the user's search request, the initial product list is obtained, which is convenient for formulating and implementing the corresponding control plan based on the current product list, instead of adjusting according to fixed rules, which helps to improve the traffic control plan. Pertinence and accuracy. In addition, the setting of the control plan is based on the control needs of the target product, which helps to intuitively meet the control needs of the product and reduces the possibility that the control plan cannot achieve the effect.

Optionally, the acquisition of the control requirements of the target commodity, and the acquisition of the target commodity based on the regulation requirements before the control scheme under the search request further includes:

Obtain historical order data of the target commodity, and obtain an order prediction model based on the historical order data;

Obtaining the exposure order data of the target product, the exposure order data at least including the historical exposure position of the target product and the order conversion data corresponding to the historical exposure position;

A conversion rate prediction model is generated based on the exposure data.

By adopting the above-mentioned technical solution, using historical order data training to obtain an order prediction model, and exposure to single data training to obtain a conversion rate prediction model, it is helpful to use the data model to assist the terminal in analyzing the order volume of different commodities and the orders of the same commodity in different exposure positions Quantity, which in turn helps to improve the output efficiency and accuracy of the control scheme.

Optionally, the acquiring historical order data of the target commodity, and acquiring an order prediction model based on the historical order data includes:

Obtain historical order data of the target commodity and at least one supplementary reference feature, the supplementary reference feature at least including holiday features and commodity quality;

An order forecasting model is obtained based on the historical order data and supplementary reference features.

By adopting the above technical solution, adding supplementary reference features when training and generating the order prediction model will help to add factors related to product sales, such as product quality and holidays, into the analysis of order prediction, which will help improve order forecasting. Predict the accuracy of the results output by the model.

Optionally, the acquisition of the control requirements of the target commodity, and the acquisition of the control scheme of the target commodity under the search request based on the regulation requirements include:

Acquiring the regulatory requirements of the target commodity;

Based on the order forecasting model and the conversion rate forecasting model, the control plan that meets the control needs is obtained;

The control scheme is used to adjust the ranking of the target commodity in the commodity list output under the same search request.

By adopting the above technical solution, using the order forecasting model and the conversion rate forecasting model in the process of generating the control plan helps to realize the generation of the control plan by using the data model, and helps to improve the generation efficiency and accuracy of the control plan.

Optionally, before generating the conversion rate prediction model based on the exposure-to-single data, the method further includes:

If the sample size of the exposed single data of the target commodity is lower than the preset sample threshold, then fitting the exposed single data to obtain a fitting exponential curve of the exposed single data;

The simulated conversion data corresponding to the exposure positions of missing exposure single data are obtained based on the fitted exponential curve.

By adopting the above technical scheme, when the amount of data generated by the model is insufficient, by fitting the exponential curve, the data that may be generated in the position of the data gap is simulated and filled, which is helpful to supplement enough data for analysis, and then Ensure that the data model is supported by a sufficient amount of data, which in turn helps to improve the accuracy of the output results of the data model.

Optionally, the secondary sorting of the initial product list based on the control scheme, and after obtaining the control product list, further includes:

Obtain the actual order data of commodities periodically according to the preset time window.

By adopting the above-mentioned technical solution, obtaining the adjusted actual order data according to the time window is helpful for analyzing the actual order data, and further helps to analyze the effect of traffic regulation through the actual order data.

Optionally, after periodically acquiring the actual order data of the target commodity according to the preset time window, the method further includes:

Obtain regulatory deviation data based on the actual order data and regulatory requirements of the target commodity;

A control compensation scheme is obtained based on the control deviation data.

By adopting the above-mentioned technical solution and obtaining deviation data through the analysis of the actual order data, it is helpful to adjust the exposure position of the product in the next time window based on the deviation data, so that the flow control of the product can meet the requirements stipulated in the control requirements. target, which in turn helps to improve the regulatory effect of the regulatory program.

In the second aspect, the present application provides a traffic control system, which adopts the following technical solution:

A flow control system, the system comprising:

An initial sorting module, configured to obtain a user's search request, and obtain an initial commodity list based on the search request, and the initial commodity list performs commodity sorting based on a preset sorting rule;

A regulation scheme module, configured to obtain regulation requirements of the target commodity, and obtain a regulation scheme of the target commodity under the search request based on the regulation requirements;

The regulation output module is used to perform secondary sorting on the initial commodity list based on the regulation scheme to obtain the regulation commodity list.

In a third aspect, the present application provides an intelligent terminal, which adopts the following technical solution:

An intelligent terminal, the intelligent terminal includes a processor and a memory, at least one instruction, at least one program, code set or instruction set are stored in the memory, the at least one instruction, the at least one program, the code The set or instruction set is loaded and executed by the processor to implement the traffic regulation method as described in any one of the first aspect.

By adopting the above technical solution, the processor in the smart terminal can implement the above flow control method according to the relevant computer program stored in the memory, and further improve the accuracy of flow control.

In a fourth aspect, the present application provides a computer-readable storage medium, which adopts the following technical solution:

A computer-readable storage medium, wherein at least one instruction, at least one program, code set or instruction set is stored in the storage medium, and the at least one instruction, the at least one program, the code set or the instruction set are processed by The device is loaded and executed to implement a traffic control method as described in any one of the first aspect.

By adopting the above technical solution, the corresponding program can be stored, thereby improving the accuracy of traffic control.

In summary, the present application includes at least one of the following beneficial technical effects:

1. After obtaining the user's search request, the initial product list is obtained, which is convenient for formulating and implementing the corresponding control plan based on the current product list, instead of adjusting according to fixed rules, which helps to improve the pertinence and accuracy of the traffic control plan In addition, the setting of the control plan is based on the control needs of the target product, which helps to intuitively meet the control needs of the product and reduces the possibility that the control plan cannot achieve the effect;

2. Use historical order data to train and obtain an order prediction model, and train exposure to single data to obtain a conversion rate prediction model, which helps the terminal analyze the order volume of different commodities and the order volume of the same commodity in different exposure positions through the data model, and then has Help to improve the output efficiency and accuracy of the control plan;

3. Obtaining deviation data through the analysis of actual order data will help to adjust the exposure position of the product in the next time window based on the deviation data, so that the flow control of the product can reach the goal stipulated in the regulation demand, and then effectively Help to improve the regulatory effect of the regulatory program.

Description of drawings

In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings that need to be used in the description of the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present invention. For those skilled in the art, other drawings can also be obtained based on these drawings without creative effort.

Fig. 1 is a method flowchart of a traffic control method shown in the embodiment of the present application;

Fig. 2 is a block flow diagram of a traffic control method shown in the embodiment of the present application;

Fig. 3 is a system block diagram of a traffic control system shown in the embodiment of the present application;

Fig. 4 is a schematic structural diagram of a smart terminal shown in an embodiment of the present application.

Explanation of reference signs: 1. Initial sorting module; 2. Regulating scheme module; 3. Regulating output module.

Detailed ways

This specific embodiment is only an explanation of this application, and it is not a limitation of this application. Those skilled in the art can make modifications to this embodiment without creative contribution according to needs after reading this specification, but as long as the rights of this application In order to make the purpose, technical solutions and advantages of the embodiments of the application clearer, the technical solutions in the embodiments of the application will be clearly described below in conjunction with the accompanying drawings 1-4 in the embodiments of the application , a complete description. Based on the embodiments in this application, all other embodiments obtained by persons of ordinary skill in the art without making creative efforts belong to the scope of protection of this application.

The processing flow shown in FIG. 1 will be described in detail below in conjunction with specific implementation methods, and the content may be as follows:

The embodiment of the present application provides a traffic control method, the method can be applied to a smart terminal, and the smart terminal is used as an execution body to control the exposure position of a product on a search result list page in an online search scenario. In addition, because the exposure position of a product on the search result list page is usually related to various factors, such as the degree of matching with the search conditions, specifically, under the same traffic control scheme, the higher the degree of matching with the search conditions, the higher the exposure. The higher the position. In order to facilitate the understanding of the solution of the present application, the search result list pages including the target commodities described in the embodiments of the present application are all generated based on the same search conditions. Based on this premise, the present application provides a traffic control method, including the following steps:

In step 101, a user's search request is obtained, and an initial product list is obtained based on the search request, and the initial product list is sorted based on a preset sorting rule.

In the implementation, before the traffic control of the product, the current traffic status of the target product must be obtained first, and the current traffic status of the product can be obtained through the current exposure position of the product. The smart terminal obtains the user's search request, and based on the user's search request, the smart terminal can obtain an initial product list matching the search request. At this time, the initial product list includes the target product, and the target product is displayed in the initial product list according to a preset sorting rule. Specifically, the preset sorting rules may be based on the correlation degree between the target product and the search request.

In this way, before traffic control is implemented, the smart terminal can obtain the initial product list to grasp the current exposure position of the target product, which in turn helps to implement a corresponding traffic control plan for the current situation of the target product.

Step 102: Obtain regulation requirements of the target commodity, and obtain regulation schemes of the target commodity under the search request based on the regulation requirements.

Specifically, the regulatory demand for the target product can have multiple demand dimensions, such as the increment of orders, the increment of followers of the store where the commodity is located, or the increment of collection of commodities, etc. This embodiment uses the increment of orders as an example to illustrate. Other situations are similar and will not be described in detail.

In the implementation, the smart terminal obtains the control requirements for the target product. After the target product generates a certain order increment, it obtains the preset control plan that matches the target product. The control plan corresponds to the target product and can target the target product itself. Product characteristics are designed. In this embodiment, the control scheme can adjust the exposure position of the target product, increase the exposure of the target product, and then realize the order growth to realize the control requirement of the target product.

In one embodiment, since the order quantity of a commodity is affected by various factors, it is difficult to analyze it through a single data, so correspondingly, the following processing can be done before step 102: obtain the historical order data of the target commodity, based on the Obtain an order prediction model for the historical order data; obtain the exposure order data of the target product, the exposure order data at least includes the historical exposure position of the target product and the order conversion data corresponding to the historical exposure position; based on the exposure cost Single data generation conversion rate prediction model.

Among them, the historical order data refers to the order data of the target product on different dates in history, with the change in date as the baseline, reflecting the change in the order volume of the target product on different days;

The exposed order data includes at least the historical exposure position of the target product and the order conversion data corresponding to the historical exposure position, taking the change of the exposure position of the product as the baseline, and reflecting the order conversion volume of the target product at different exposure positions.

In the implementation, the conversion rate refers to the probability that the user will generate an actual order after clicking on the product. The smart terminal can use the exposed single data as the training data set and train the ESMM (Entire Space Multi Task) twin-tower model to achieve this. The data used in the ESMM twin-tower model can include user search click logs, user click data, and corresponding order information. In addition, it can also include the basic information of the target product, including product price, product use, store address, etc.; user behavior Data, including data in dimensions such as the number of orders placed, favorites, or reposts after the user clicks to browse the product.

In this way, the ESMM twin-tower model shows the introduction of CTR (Click-Through-Rate, click-through rate) and CTCVR (Click-Through-Conversion- Rate, click and convert) is used as an auxiliary task to learn CVR (Conversion Rate, conversion rate) in a detour, so as to realize the training and prediction work of the conversion rate prediction model. In addition, the AUC value can also be used to evaluate the effect of the conversion rate prediction model.

In one embodiment, when the amount of historical exposure data is small and only part of the exposure data can be obtained, the accuracy of the model calculation results may be reduced. Therefore, correspondingly, the above conversion rate The following processing may also be included during the prediction model training: if the sample size of the exposure data of the target commodity is lower than the preset sample threshold, then fitting the exposure data to obtain the fitting exponential curve of the exposure data; The fitting exponential curve obtains the simulated transformation data corresponding to the exposure positions where the single exposure data is missing.

In the implementation, after the smart terminal obtains the data associated with the target commodity for training the conversion rate prediction model, it can compare the obtained sample size with the preset sample threshold. If the sample size is lower than the sample threshold, it needs to The data can be supplemented, and the simulated transformation data can be obtained at the exposure position where the missing exposure becomes single data by means of exponential curve fitting. Specifically, the sample threshold can be set based on the exposure position, and the sample threshold can be set to 20, 25, or 30, etc., that is, the smart terminal needs to at least obtain the target product at the exposure position of 1-20, 1-25, 1-30 All exposures in between are single data.

In implementation, when the smart terminal fits the exponential curve, the exposure position of the target product can be used as the X axis, and the conversion rate corresponding to the target product at different exposure positions is the Y axis, and ^R2 can be 0.9801. In this way, the exponential curve can be made close to the relationship between the average exposure position and the conversion rate, and the formula can be as follows:

y=0.004ln(x)+0.0193

R2 = 0.9801

In this way, during the training process of the conversion rate prediction model, missing data can be simulated and supplemented, which helps to improve the accuracy of the conversion rate prediction model.

On the other hand, the smart terminal can obtain the historical order data of the target product and the basic information of the target product itself, including product price, product use, store address, etc., and use this as a training data set to train and obtain order predictions. Model. Since order forecasting has high requirements on the update performance of the forecasting model, the order forecasting model can be obtained by training with the LightGBM model, and can be forecasted in a rolling iterative manner, that is, the data within a certain time window can be selected as the training data set to predict the next The order volume in a time window to be predicted, as time rolls, the order volume in the next time window to be predicted is reconstructed into the training data set, and then forecasted again. In this embodiment, rolling iterative training can be performed with one week as the time window.

In one embodiment, the order quantity of the target commodity may be affected by various factors, so the training process of the order prediction model may include the following processing: obtaining the historical order data of the target commodity and at least one supplementary reference feature, The supplementary reference features include at least holiday features and commodity quality; an order prediction model is obtained based on the historical order data and supplementary reference features.

In implementation, when the smart terminal builds the training data set of the order prediction model, it can add supplementary reference features to the training data set. The supplementary reference features can be set from multiple dimensions associated with the order quantity of the target product, for example: target product Quality fluctuations, time window adjustments for data statistics, product order fluctuation data and corresponding fluctuation factors, historical order volume and corresponding order time, holidays and other characteristic dimensions.

In this way, the training data set of the order prediction model built by the smart terminal can be based on historical order data, supplemented with the basic information of the target product itself and supplementary reference features, which is conducive to increasing the accuracy of order prediction.

In one embodiment, after obtaining the order prediction model and the conversion rate prediction model, the following processing can be performed: obtain the control requirements of the target commodity; obtain a control plan that meets the control requirements based on the order prediction model and the conversion rate prediction model; The regulation scheme is used to perform ranking adjustment of the target product in the product list output under the same search request.

In implementation, after the smart terminal obtains the user's control needs, it can generate a corresponding control plan through the order prediction model and the conversion rate prediction model. Taking the control requirement mentioned above in this embodiment as an example to generate an order increment A for the target commodity within the next time window, when the smart terminal obtains the regulation requirement to generate an order increment A, it first obtains the target commodity currently in the default order. Exposure position, the smart terminal can use the order prediction model to predict the order volume of the current target product in the next time window. Assuming that the forecast result is that the default order volume is B, the smart terminal can obtain the final demand for the target product in the next time window. The order volume is A+B. At this time, the smart terminal can use the conversion rate prediction model to predict the conversion rate of the product at different exposure positions, and how many exposure positions need to be increased to obtain the order volume of A+B.

In this way, the smart terminal can obtain a control scheme for the control requirements of the target commodity, that is, an adjustment scheme for exposure positions in different time windows. And then precisely control the regulation and control demand of the target commodity.

Step 103: Perform secondary sorting on the initial product list based on the control scheme to obtain a list of regulated products.

In the implementation, referring to Figure 2, after the intelligent terminal generates the control plan through the order prediction model and the conversion rate prediction model, it can implement the control plan for the target product. It should be noted that due to the different time windows, the control plan can include one or more The adjustment of the position of the second exposure, so that the total number of order increments obtained by the target commodity in one or more time windows reaches the increment required in the regulation demand. For example, the regulatory demand of the target product involves three time windows, which are time windows 1, 2, and 3. The regulatory demand of the target product is order increment A. In time window 1, the target product needs to be adjusted to the fourth place; in the time window The target product needs to be adjusted to the 6th position within 2; the target product needs to be adjusted to the 7th position within the time window 3, and the order increments of a, b, and c are respectively reached within the three time windows, then when A=a When +b+c, the control requirements are met.

In this way, the smart terminal assists the smart terminal to generate a control plan by training the order prediction model and the conversion rate prediction model obtained from the product-related data, and guides the adjustment of the exposure position of the product, so that the smart terminal obtains the regulated product list after traffic control. When the user enters a specified search request, the smart terminal will return the list of regulated products to the user, so that the regulation needs of the target product can be realized.

In one embodiment, the adjustment plan involving multiple exposure position adjustments spans multiple time windows, which may lead to inaccurate predictions, thereby reducing the accuracy of traffic control. Therefore, correspondingly, the following processing can be performed after step 103: according to the preset The set time window periodically obtains the actual order data of the product, and obtains the control deviation data based on the actual order data of the target product and the control demand; obtains the control compensation plan based on the control deviation data.

In the implementation, the smart terminal can obtain the actual order data Xi in the current time window and the difference ΔXi between the target order and the actual order of the previous day, and set the initial control coefficient αi. Through the above data, each time window can be calculated The update value α of the control coefficient in , the formula is as follows:

α=αi*(Xi+ΔXi)/Xi

In this way, the smart terminal can update the update value of the control coefficient in the next time window, and the smart terminal can update the control scheme according to the update value, so that the control requirements can be met.

Based on the same technical concept, the embodiment of the present invention also provides a traffic control system, referring to Figure 3, the system includes:

The initial sorting module 1 is used to obtain a user's search request, and obtain an initial commodity list based on the search request, and the initial commodity list performs commodity sorting based on a preset sorting rule;

The regulation scheme module 2 is used to obtain the regulation requirements of the target commodity, and obtain the regulation scheme of the target commodity under the search request based on the regulation requirements;

The regulation output module 3 is configured to perform secondary sorting on the initial commodity list based on the regulation scheme, and obtain the regulation commodity list.

Optionally, the system also includes:

An order forecasting model module, configured to acquire historical order data of the target commodity, and acquire an order forecasting model based on the historical order data;

The conversion rate data module is used to obtain the exposure order data of the target product, and the exposure order data at least includes the historical exposure position of the target product and the order conversion data corresponding to the historical exposure position;

A conversion rate prediction model module, configured to generate a conversion rate prediction model based on the exposure data.

Optionally, the order forecasting model module includes:

The supplementary feature submodule is used to obtain the historical order data of the target commodity and at least one supplementary reference feature, the supplementary reference feature includes at least holiday features and commodity quality;

The model generation submodule is used to obtain an order forecasting model based on the historical order data and supplementary reference features.

Optionally, the control scheme module 2 includes:

A demand acquisition module, configured to acquire the regulatory demand of the target commodity;

The scheme generation module is used to obtain a control scheme that meets the control requirements based on the order prediction model and the conversion rate prediction model; the control scheme is used to implement the ranking adjustment of the target commodity in the commodity list output under the same search request.

Optionally, the conversion rate prediction model module includes:

The data fitting sub-module is used to fit the exposure data to obtain the fitting exponential curve of the exposure data if the sample size of the exposure data of the target commodity is lower than the preset sample threshold;

The data filling sub-module is used to obtain, based on the fitting exponential curve, the simulated conversion data corresponding to the exposure positions where the missing exposures become single data.

Optionally, after regulating the output module 3, it also includes:

The actual data acquisition module is used to periodically acquire the actual order data of the commodity according to the preset time window;

The deviation acquisition module is used to obtain control deviation data based on the actual order data and control requirements of the target commodity;

A regulation compensation module, configured to obtain a regulation compensation scheme based on the regulation deviation data.

The embodiment of the present application also discloses a smart terminal. Referring to FIG. 4 , the smart terminal includes a memory and a processor, and the memory stores a computer program that can be loaded by the processor and execute the traffic control method as described above.

Based on the same technical concept, the embodiment of the present application also discloses a computer-readable storage medium, which includes various steps in the flow of the above-mentioned traffic control method that can be implemented when loaded and executed by a processor.

The computer-readable storage medium includes, for example, a USB flash drive, a mobile hard disk, a read-only memory (Read-Only Memory, ROM), a random access memory (Random Access Memory, RAM), a magnetic disk or an optical disk, etc., which can store program codes. medium.

Those skilled in the art can clearly understand that for the convenience and simplification of description, only the division of the above-mentioned functional modules is used as an example for illustration. In practical applications, the above-mentioned function allocation can be completed by different functional modules according to needs. The internal structure of the device is divided into different functional modules to complete all or part of the functions described above. For the specific working process of the system, device and unit described above, you can refer to the corresponding process in the foregoing method embodiment, and will not repeat them here. .

The above embodiments are only used to introduce the technical solutions of the present application in detail, but the descriptions of the above embodiments are only used to help understand the method and core idea of the present application, and should not be construed as limiting the present application. Within the technical scope disclosed in this application, changes or substitutions that can be easily conceived by those skilled in the art shall fall within the protection scope of this application.

Claims

A flow control method, characterized in that the method comprises the following steps:

Obtain a user's search request, and obtain an initial product list based on the search request, and the initial product list is based on a preset sorting rule for product sorting;

Acquiring the regulation requirements of the target commodity, and obtaining the regulation plan of the target commodity under the search request based on the regulation requirements;

Based on the control scheme, the initial product list is sorted twice to obtain the list of regulated products.
A traffic control method according to claim 1, characterized in that: said obtaining the control requirements of the target commodity, and obtaining the control plan of the target commodity based on the regulation requirements before the search request further includes:

Obtain historical order data of the target commodity, and obtain an order prediction model based on the historical order data;

Obtaining the exposure order data of the target product, the exposure order data at least including the historical exposure position of the target product and the order conversion data corresponding to the historical exposure position;

A conversion rate prediction model is generated based on the exposure data.
A traffic control method according to claim 2, characterized in that: said acquiring historical order data of said target commodity, and acquiring an order prediction model based on said historical order data comprises:

Obtain historical order data of the target commodity and at least one supplementary reference feature, the supplementary reference feature at least including holiday features and commodity quality;

An order forecasting model is obtained based on the historical order data and supplementary reference features.
A traffic regulation method according to claim 3, characterized in that: the acquisition of the regulation requirements of the target commodity, and the regulation and control scheme of the target commodity under the search request based on the regulation requirements include:

Acquiring the regulatory requirements of the target commodity;

Based on the order forecasting model and the conversion rate forecasting model, the control plan that meets the control needs is obtained;

The control scheme is used to adjust the ranking of the target commodity in the commodity list output under the same search request.
A traffic control method according to claim 3, characterized in that: before generating the conversion rate prediction model based on the exposure into single data, it also includes:

If the sample size of the exposed single data of the target commodity is lower than the preset sample threshold, then fitting the exposed single data to obtain a fitting exponential curve of the exposed single data;

The simulated conversion data corresponding to the exposure positions of missing exposure single data are obtained based on the fitted exponential curve.
A traffic control method according to claim 1, characterized in that: performing secondary sorting on the initial product list based on the control plan, and after obtaining the control product list, further includes:

Obtain the actual order data of commodities periodically according to the preset time window.
A traffic control method according to claim 6, characterized in that: after periodically acquiring the actual order data of the target commodity according to the preset time window, it further includes:

Obtain regulatory deviation data based on the actual order data and regulatory requirements of the target commodity;

A control compensation scheme is obtained based on the control deviation data.
A flow control system, characterized in that the system comprises:

An initial sorting module (1), configured to obtain a user's search request, and obtain an initial commodity list based on the search request, and the initial commodity list performs commodity sorting based on a preset sorting rule;

A regulation plan module (2), configured to obtain the regulation requirements of the target commodity, and obtain the regulation plan of the target commodity under the search request based on the regulation requirements;

A regulation output module (3), configured to perform secondary sorting on the initial commodity list based on the regulation scheme to obtain the regulation commodity list.
An intelligent terminal, characterized in that the intelligent terminal includes a processor and a memory, at least one instruction, at least one program, code set or instruction set are stored in the memory, and the at least one instruction, the at least one program , the code set or instruction set is loaded and executed by the processor to implement a traffic regulation method as claimed in any one of claims 1 to 7.
A computer-readable storage medium, characterized in that at least one instruction, at least one program, code set or instruction set is stored in the storage medium, and the at least one instruction, the at least one program, the code set or The instruction set is loaded and executed by the processor to implement a traffic control method according to any one of claims 1 to 7.