CN113408797B

CN113408797B - Method for generating multi-time sequence model of flow quantity prediction, method and device for sending information

Info

Publication number: CN113408797B
Application number: CN202110631893.XA
Authority: CN
Inventors: 王应德; 庄晓天
Original assignee: Beijing Jingdong Zhenshi Information Technology Co Ltd
Current assignee: Beijing Jingdong Zhenshi Information Technology Co Ltd
Priority date: 2021-06-07
Filing date: 2021-06-07
Publication date: 2023-09-26
Anticipated expiration: 2041-06-07
Also published as: CN113408797A

Abstract

The embodiment of the invention discloses a method for generating a circulation quantity prediction multi-time sequence model, and a method and a device for sending information. One embodiment of the method comprises the following steps: acquiring a historical circulation quantity set of a target object in a preset time period; inputting the historical circulation quantity set into each circulation quantity prediction time sequence model to obtain a predicted circulation quantity set; determining each historical circulation quantity of each return time granularity in a predicted circulation quantity group set and a historical circulation quantity set as an input parameter of a preset linearization objective function so as to generate an objective function to be solved; based on the constraint condition set, solving the objective function to be solved to obtain a model weight coefficient set; and carrying out weighted combination processing on the circulation quantity prediction time sequence model corresponding to each model weight coefficient in the model weight coefficient set according to the model weight coefficient set to obtain a circulation quantity prediction multi-time sequence model. The method and the device improve accuracy, robustness and stability of the flow quantity prediction, and simplify a model determining process.

Description

Method for generating multi-time sequence model of flow quantity prediction, method and device for sending information

Technical Field

Embodiments of the present disclosure relate to the field of computer technology, and in particular, to a method for generating a flow quantity prediction multi-timing model, an information sending method, an information sending device, an electronic device, and a computer readable medium.

Background

Demand Forecast (Demand Forecast) is a very important ring in the field of supply chains, and performs three lines of defense, collectively referred to as supply chains, with inventory planning and supply chains. Currently, in generating a predicted amount of flow, the following methods are generally adopted: the historical flow volume data and the selected one or more time sequence models are used for generating the flow volume of a future period of time.

However, when the predicted flow amount is generated in the above manner, there are often the following technical problems: when a single time sequence model is adopted, the multi-periodic characteristics of the time sequence cannot be covered, so that the prediction accuracy and the robustness are poor; when a plurality of time sequence models are adopted, each time sequence model needs to be selected in advance, the number of the time sequence models and the weight of each time sequence model are determined, the stability of a prediction result is poor, and the model determination process is complicated.

Disclosure of Invention

The disclosure is in part intended to introduce concepts in a simplified form that are further described below in the detailed description. The disclosure is not intended to identify key features or essential features of the claimed subject matter, nor is it intended to be used to limit the scope of the claimed subject matter.

Some embodiments of the present disclosure propose a flow amount prediction multi-timing model generation method, an information transmission method, an apparatus, an electronic device, and a computer readable medium to solve one or more of the technical problems mentioned in the background section above.

In a first aspect, some embodiments of the present disclosure provide a method for generating a flow amount prediction multi-timing model, the method including: acquiring a historical circulation quantity set of a target object in a preset time period; inputting the historical circulation quantity set into each circulation quantity prediction time sequence model in a circulation quantity prediction time sequence model pool, and obtaining a prediction circulation quantity of each return time granularity of the target object in a preset return time period as a prediction circulation quantity group to form a prediction circulation quantity group set, wherein the preset time period comprises at least one periodic return time period of the preset return time period; determining each historical circulation quantity of each return time granularity in the predicted circulation quantity group set and the historical circulation quantity set as an input parameter of a preset linearization objective function so as to generate an objective function to be solved; solving the objective function to be solved based on a constraint condition set corresponding to the preset linearization objective function to obtain a model weight coefficient set, wherein the model weight coefficient in the model weight coefficient set corresponds to the flow quantity prediction time sequence model in the flow quantity prediction time sequence model pool, and the constraint condition set comprises constraint conditions for representing constraint on the number of models; and carrying out weighted combination processing on the circulation quantity prediction time sequence model corresponding to each model weight coefficient in the model weight coefficient set according to the model weight coefficient set to obtain a circulation quantity prediction multi-time sequence model.

Optionally, before the solving process is performed on the objective function to be solved, the method further includes: generating model errors corresponding to the flow quantity prediction time sequence model based on the predicted flow quantity group corresponding to each flow quantity prediction time sequence model in the flow quantity prediction time sequence model pool and each historical flow quantity of each return time granularity in the historical flow quantity set, and obtaining a model error set.

Optionally, after generating the model error corresponding to the flow quantity prediction time sequence model to obtain a model error set, the method further includes: deleting each constraint condition of the constraint condition set characterization, which is used for constraining the model weight coefficient, so as to update the constraint condition set.

Optionally, after deleting each constraint condition in the constraint condition set for characterizing the constraint on the model weight coefficient, the method further includes: determining the model error set as input parameters of a first model weight function, a second model weight function and a third model weight function to generate a first model weight function to be weighted, a second model weight function to be weighted and a third model weight function to be weighted; weighting the first model weight function to be weighted, the second model weight function to be weighted and the third model weight function to be weighted to obtain a weighted model weight function as a weight constraint condition; the weight constraints are added to the constraint set to update the constraint set.

In a second aspect, some embodiments of the present disclosure provide an information transmission method, the method including: acquiring a historical circulation quantity set of a target object in a preset time period; inputting the historical circulation quantity set into a circulation quantity prediction multi-time sequence model to obtain a predicted circulation quantity of each prediction time granularity of the target object in a preset prediction time period as a target predicted circulation quantity; and sending the obtained target predicted stream quantity to an associated display device.

Optionally, the acquiring the historical circulation quantity set of the target object in the preset time period further includes: and acquiring the stock quantity of the target object.

Optionally, the method further comprises: and controlling the associated dispatching equipment to execute dispatching operation according to the stock quantity and the obtained target predicted flow quantity in response to the stock quantity and the obtained target predicted flow quantity meeting preset replenishment conditions.

In a third aspect, some embodiments of the present disclosure provide a flow amount prediction multi-timing model generating apparatus, the apparatus including: an acquisition unit configured to acquire a set of historical circulation amounts of a target article within a preset time period; an input unit configured to input the historical circulation quantity set to each circulation quantity prediction time sequence model in a circulation quantity prediction time sequence model pool, and obtain a prediction circulation quantity of each return time granularity of the target object in a preset return time period as a prediction circulation quantity group to form a prediction circulation quantity group set, wherein the preset time period comprises at least one periodic return time period of the preset return time period; a determining unit configured to determine each historical circulation quantity of each return time granularity in the predicted circulation quantity set and the historical circulation quantity set as an input parameter of a preset linearization objective function so as to generate an objective function to be solved; a generating unit configured to perform solving processing on the objective function to be solved based on a constraint condition set corresponding to the preset linearization objective function, so as to obtain a model weight coefficient set, wherein a model weight coefficient in the model weight coefficient set corresponds to a flow volume prediction time sequence model in the flow volume prediction time sequence model pool, and the constraint condition set comprises a constraint condition for representing constraint on the number of models; and the weighted combination unit is configured to perform weighted combination processing on the circulation quantity prediction time sequence model corresponding to each model weight coefficient in the model weight coefficient set according to the model weight coefficient set to obtain a circulation quantity prediction multi-time sequence model.

Optionally, before generating the unit, the apparatus further comprises: and a model error generating unit configured to generate a model error corresponding to the flow amount prediction time sequence model based on the predicted flow amount group corresponding to each flow amount prediction time sequence model in the flow amount prediction time sequence model pool in the predicted flow amount group and each historical flow amount of each return time granularity in the historical flow amount set, and obtain a model error set.

Optionally, after the model error generation unit, the apparatus further comprises: and the deleting unit is configured to delete each constraint condition of the constraint condition set characterization, which is used for constraining the model weight coefficient, so as to update the constraint condition set.

Optionally, after deleting the unit, the apparatus further comprises: the device comprises an input parameter determining unit, a weighting processing unit and an adding unit. Wherein the input parameter determination unit is configured to determine the above-mentioned set of model errors as input parameters of the first model weight function, the second model weight function and the third model weight function to generate a first model weight function to be weighted, a second model weight function to be weighted and a third model weight function to be weighted. The weighting processing unit is configured to perform weighting processing on the first model weight function to be weighted, the second model weight function to be weighted and the third model weight function to be weighted, so as to obtain a weighted model weight function as a weight constraint condition. The adding unit is configured to add the above weight constraint to the constraint set to update the constraint set. In a fourth aspect, some embodiments of the present disclosure provide an information transmitting apparatus, the apparatus including: a history circulation quantity acquisition unit configured to acquire a history circulation quantity set of a target article in a preset time period; a history circulation quantity input unit configured to input the history circulation quantity set into a circulation quantity prediction multi-time sequence model, and obtain a predicted circulation quantity of each prediction time granularity of the target object in a preset prediction time period as a target predicted circulation quantity; and a transmitting unit configured to transmit the obtained target predicted stream quantity to an associated display device.

Optionally, the historical flow amount set obtaining unit further includes: and an inventory amount acquisition unit configured to acquire an inventory amount of the target article.

Optionally, the apparatus further comprises: and a control unit configured to control the associated scheduling device to perform a scheduling operation according to the inventory amount and the obtained target predicted throughput in response to the inventory amount and the obtained target predicted throughput satisfying a preset replenishment condition.

In a fifth aspect, some embodiments of the present disclosure provide an electronic device comprising: one or more processors; and a storage device having one or more programs stored thereon, which when executed by the one or more processors, cause the one or more processors to implement the method described in any of the first or second implementations.

In a sixth aspect, some embodiments of the present disclosure provide a computer readable medium having a computer program stored thereon, wherein the program when executed by a processor implements the method described in any of the above first or second aspects.

The above embodiments of the present disclosure have the following advantageous effects: according to the circulation quantity prediction multi-time sequence model obtained by the circulation quantity prediction multi-time sequence model generation method, the prediction accuracy, the robustness and the stability of a prediction result are improved, and the model determination process is simplified. Specifically, the reason why the prediction accuracy, the robustness and the stability of the prediction result are poor and the model determination process is complicated is that: when a single time sequence model is adopted, the multi-periodic characteristics of the time sequence cannot be covered, so that the prediction accuracy and the robustness are poor; when a plurality of time sequence models are adopted, each time sequence model needs to be selected in advance, the number of the time sequence models and the weight of each time sequence model are determined, the stability of a prediction result is poor, and the model determination process is complicated. Based on this, the method for generating the multi-timing model for predicting the circulation quantity according to some embodiments of the present disclosure first obtains a set of historical circulation quantities of the target article in a preset time period. And then, inputting the historical circulation quantity set into each circulation quantity prediction time sequence model in a circulation quantity prediction time sequence model pool, and obtaining the predicted circulation quantity of each return time granularity of the target object in a preset return time period as a predicted circulation quantity group so as to form a predicted circulation quantity group set. The preset time period comprises at least one periodic time period of the preset time period. Therefore, the circulation quantity of the target object at each return time granularity can be predicted through each circulation quantity prediction time sequence model in the circulation quantity prediction time sequence model pool, a predicted circulation quantity group set obtained through prediction is used as an input parameter, and the method can be used for solving the weight of the target function to be solved. And then, determining each historical circulation quantity of each return time granularity in the predicted circulation quantity group set and the historical circulation quantity set as an input parameter of a preset linearization objective function so as to generate an objective function to be solved. Thus, the objective function to be solved can be used as a target condition when determining the model weight coefficient, so that the determined model weight coefficient is the optimal solution. And secondly, solving the objective function to be solved based on a constraint condition set corresponding to the preset linearization objective function to obtain a model weight coefficient set. Wherein the model weight coefficients in the model weight coefficient set correspond to the flow volume prediction time sequence model in the flow volume prediction time sequence model pool, and the constraint condition set comprises constraint conditions for representing the constraint on the number of the models. Therefore, the model weight coefficient set obtained by solving processing can be the optimal solution, and each constraint condition in the constraint condition sets can be met. And finally, carrying out weighted combination processing on the flow quantity prediction time sequence model corresponding to each model weight coefficient in the model weight coefficient set according to the model weight coefficient set to obtain a flow quantity prediction multi-time sequence model. Therefore, the multi-time sequence model for predicting the circulation quantity of the corresponding target object can be obtained by weighting and combining according to the optimal model weight coefficient set. And when the circulation quantity prediction multi-time sequence model is generated, the prediction circulation quantity of the target object in the preset return time period is determined according to the historical circulation quantity set of the target object in the preset time period, so that the multi-periodicity characteristic of the circulation quantity of the target object in the historical time period can be covered, and the prediction accuracy is improved. And because the multi-time sequence model of the circulation quantity prediction is generated aiming at the target object, the influence of the circulation quantity change characteristics of different objects in the history time (such as the influence of larger circulation quantity difference of different objects in different seasons) is reduced, and the robustness of the prediction is improved. And because the selection of the model and the determination of the model weight coefficient are obtained through solving, whether the model is selected or not can be represented by the model weight coefficient, so that the number of each time sequence model and the weight of each time sequence model do not need to be selected in advance, the stability of a prediction result is improved, and the model determination process is simplified.

Drawings

The above and other features, advantages, and aspects of embodiments of the present disclosure will become more apparent by reference to the following detailed description when taken in conjunction with the accompanying drawings. The same or similar reference numbers will be used throughout the drawings to refer to the same or like elements. It should be understood that the figures are schematic and that elements and components are not necessarily drawn to scale.

FIG. 1 is a schematic illustration of one application scenario of a flow amount prediction multi-timing model generation method according to some embodiments of the present disclosure;

fig. 2 is a schematic diagram of one application scenario of an information transmission method according to some embodiments of the present disclosure;

FIG. 3 is a flow chart of some embodiments of a method of generating a flow quantity prediction multi-timing model according to the present disclosure;

FIG. 4 is a flow chart of some embodiments of an information transmission method according to the present disclosure;

FIG. 5 is a schematic diagram of some embodiments of a flow amount prediction multi-temporal model generation device according to the present disclosure;

fig. 6 is a schematic structural diagram of some embodiments of an information transmission apparatus according to the present disclosure;

fig. 7 is a schematic structural diagram of an electronic device suitable for use in implementing some embodiments of the present disclosure.

Detailed Description

Embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. While certain embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be embodied in various forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete. It should be understood that the drawings and embodiments of the present disclosure are for illustration purposes only and are not intended to limit the scope of the present disclosure.

It should be noted that, for convenience of description, only the portions related to the present invention are shown in the drawings. Embodiments of the present disclosure and features of embodiments may be combined with each other without conflict.

It should be noted that the terms "first," "second," and the like in this disclosure are merely used to distinguish between different devices, modules, or units and are not used to define an order or interdependence of functions performed by the devices, modules, or units.

It should be noted that references to "one", "a plurality" and "a plurality" in this disclosure are intended to be illustrative rather than limiting, and those of ordinary skill in the art will appreciate that "one or more" is intended to be understood as "one or more" unless the context clearly indicates otherwise.

The names of messages or information interacted between the various devices in the embodiments of the present disclosure are for illustrative purposes only and are not intended to limit the scope of such messages or information.

The present disclosure will be described in detail below with reference to the accompanying drawings in conjunction with embodiments.

Fig. 1 is a schematic diagram of an application scenario of a method for generating a flow amount prediction multi-timing model according to some embodiments of the present disclosure.

In the application scenario of fig. 1, first, the computing device 101 may obtain a set of historical circulation amounts 102 of the target item over a preset period of time. Then, the computing device 101 may input the historical circulation quantity set 102 into each circulation quantity prediction time sequence model in the circulation quantity prediction time sequence model pool 103, so as to obtain a predicted circulation quantity of each return time granularity of the target article in a preset return time period as a predicted circulation quantity group to form a predicted circulation quantity group set 104. The preset time period comprises at least one periodic time period of the preset time period. Thereafter, the computing device 101 may determine each historical flow amount 105 of each return time granularity in the set of predicted flow amounts 104 and the set of historical flow amounts 102 as an input parameter of a preset linearization objective function 106, so as to generate an objective function 107 to be solved. Next, the computing device 101 may perform a solving process on the objective function 107 to be solved based on the constraint condition set 108 corresponding to the preset linearized objective function 106, to obtain a model weight coefficient set 109. Wherein the model weight coefficients in the model weight coefficient set 109 correspond to the flow amount prediction time series model in the flow amount prediction time series model pool 103. The set of constraints 108 includes constraints that characterize constraints on the number of models. Finally, the computing device 101 may perform weighted combination processing on the flow amount prediction time sequence model corresponding to each model weight coefficient in the model weight coefficient set 109 according to the model weight coefficient set 109, to obtain a flow amount prediction multi-time sequence model 110.

The computing device 101 may be hardware or software. When the computing device is hardware, the computing device may be implemented as a distributed cluster formed by a plurality of servers or terminal devices, or may be implemented as a single server or a single terminal device. When the computing device is embodied as software, it may be installed in the hardware devices listed above. It may be implemented as a plurality of software or software modules, for example, for providing distributed services, or as a single software or software module. The present invention is not particularly limited herein.

It should be understood that the number of computing devices in fig. 1 is merely illustrative. There may be any number of computing devices, as desired for an implementation.

Fig. 2 is a schematic diagram of an application scenario of an information transmission method according to some embodiments of the present disclosure.

In the application scenario of fig. 2, first, the computing device 201 may obtain a set 202 of historical amounts of flow of the target item over a preset period of time. Then, the computing device 201 may input the historical circulation quantity set 202 into the circulation quantity prediction multi-timing model 203, so as to obtain a predicted circulation quantity of each prediction time granularity of the target item within a preset prediction time period as a target predicted circulation quantity. For example, the flow amount prediction multi-timing model 203 may be the flow amount prediction multi-timing model 110 in fig. 1. Finally, the computing device 201 may send the resulting target predicted traffic 204 to the associated display device 205.

The computing device 201 may be hardware or software. When the computing device is hardware, the computing device may be implemented as a distributed cluster formed by a plurality of servers or terminal devices, or may be implemented as a single server or a single terminal device. When the computing device is embodied as software, it may be installed in the hardware devices listed above. It may be implemented as a plurality of software or software modules, for example, for providing distributed services, or as a single software or software module. The present invention is not particularly limited herein.

It should be understood that the number of computing devices in fig. 2 is merely illustrative. There may be any number of computing devices, as desired for an implementation.

The execution subjects of the method for generating the multi-time sequence model of the flow amount prediction and the method for transmitting the information may be the same computing device or may be different computing devices.

With continued reference to fig. 3, a flow 300 of some embodiments of a flow quantity prediction multi-timing model generation method according to the present disclosure is shown. The method for generating the flow quantity prediction multi-time sequence model comprises the following steps:

step 301, acquiring a historical circulation quantity set of a target object in a preset time period.

In some embodiments, the execution subject of the method for generating a multi-timing model for predicting the circulation amount (e.g., the computing device 101 shown in fig. 1 or the computing device 201 shown in fig. 2) may obtain the set of the historical circulation amounts of the target item in the preset period from the terminal storing the historical circulation amounts of the target item through a wired connection manner or a wireless connection manner. In practice, the executing body may obtain, from the terminal, a historical circulation quantity of which the corresponding article identifier is the same as the article identifier of the target article and the circulation time is in the preset time period, so as to obtain a historical circulation quantity set. Wherein the target item may be a currently selected item. The historical circulation amount in the historical circulation amount set may be a circulation amount of the target item at a historical time (for example, the historical circulation amount may be a sales amount of the target item at the historical time). The article identifier may be used to uniquely identify an article. The above-described preset time period may be a preset time period including each cycle time period. For example, the predetermined time period may be 2018/1/1-2020/12/31. The time granularity of the preset time period can be 2018/1/1-2020/12/31 each day. The preset time period may include a period of time from 2018/1/1 to 2019/1/1, from 2019/1/1 to 2020/1/1. It should be noted that the wireless connection may include, but is not limited to, 3G/4G connections, wiFi connections, bluetooth connections, wiMAX connections, zigbee connections, UWB (ultra wideband) connections, and other now known or later developed wireless connection means.

Step 302, the historical circulation quantity set is input to each circulation quantity prediction time sequence model in the circulation quantity prediction time sequence model pool, and the prediction circulation quantity of each return time granularity of the target object in the preset return time period is obtained and used as a prediction circulation quantity group to form a prediction circulation quantity group set.

In some embodiments, the execution body may input the historical circulation quantity set into each circulation quantity prediction time sequence model in the circulation quantity prediction time sequence model pool, so as to obtain a predicted circulation quantity of each return time granularity of the target article in a preset return time period as a predicted circulation quantity group to form a predicted circulation quantity group set. Wherein each predicted stream quantity in the predicted stream quantity group corresponds to each return time granularity. Each predicted circulation quantity group in the predicted circulation quantity group corresponds to each circulation quantity prediction time sequence model. The flow quantity prediction time sequence model pool can be a set formed by each flow quantity prediction time sequence model. The flow amount prediction timing model may be a timing model for predicting the flow amount of the article. For example, the above-described pool of flow rate prediction timing models may include, but are not limited to, the following timing models for predicting the flow rate of an item: ETS (exponential smoothing algorithm), hot-Winter (third-order exponential smoothing algorithm), SES (simple exponential smoothing algorithm), SARIMA (autoregressive differential moving average algorithm with seasonality), SA (simple average algorithm). The preset return time period may be a preset time period for returning the circulation amount of the target article within the preset time period. The preset time period includes at least one periodic time period of the preset time period. The period of time may be a period of time corresponding to the preset period of time. For example, the preset return time period may be 2020/12/1-2020/12/31. The period of time for the cycle back may be 2018/12/1-2018/12/31 and 2019/12/1-2019/12/31. The return time granularity among the respective return time granularities may be a unit time for further dividing the preset return time period. For example, the individual return time granularity described above may be for each day within the preset return time period 2020/12/1-2020/12/31. Therefore, the circulation quantity of the target object at each return time granularity can be predicted through each circulation quantity prediction time sequence model in the circulation quantity prediction time sequence model pool, a predicted circulation quantity group set obtained through prediction is used as an input parameter, and the method can be used for solving the weight of the target function to be solved.

And 303, determining each historical circulation quantity of each return time granularity in the predicted circulation quantity group set and the historical circulation quantity set as an input parameter of a preset linearization objective function so as to generate an objective function to be solved.

In some embodiments, the execution body may determine each historical flow amount of each return time granularity in the set of predicted flow amounts and the set of historical flow amounts as an input parameter of a preset linearization objective function, so as to generate an objective function to be solved. The preset linearization objective function may be an objective function obtained by linearizing the preset objective function. For example, the objective function may be a function that minimizes the value of MAPE (Mean AbsolutePercentage Error, average absolute percentage error), and may be expressed as:

wherein H represents the number of the granularity of the return time included in the preset return time period. t represents the sequence number of the return time granularity. N represents the number of the flow amount prediction time sequence models included in the flow amount prediction time sequence model pool. i represents the sequence number of the flow quantity prediction sequence model in the flow quantity prediction sequence model pool. Omega _i And the model weight coefficient of the ith circulation quantity prediction time sequence model is represented. Representing the predicted amount of flow of the target item at the granularity of the t-th return time, generated by the i-th flow amount prediction timing model. y is ^his t represents the historical circulation quantity of the target object at the time granularity of the t-th return.

The preset linearization objective function, that is, the objective function after linearization may be:

wherein z is _t Representation of

The generated objective function to be solved may be used as a target condition when determining the model weight coefficient, so that the determined model weight coefficient is the optimal solution, via step 303.

And step 304, carrying out solving processing on the objective function to be solved based on a constraint condition set corresponding to the preset linearization objective function to obtain a model weight coefficient set.

In some embodiments, the executing body may perform a solving process on the objective function to be solved based on a constraint condition set corresponding to the preset linearization objective function, to obtain a model weight coefficient set. The constraint condition set may be a condition for constraining the numerical value of the coefficient in the preset linearization objective function. The set of constraints includes constraints that characterize constraints on the number of models. The model weight coefficients in the model weight coefficient set correspond to the flow amount prediction time sequence model in the flow amount prediction time sequence model pool. For example, the constraints included in the constraint set described above may be expressed as the following mathematical model:

ω _i ∈[0，1] (1)

Wherein Q is an arbitrarily large integer.

Where P represents the number threshold of the selected flow amount prediction timing model. P is less than or equal to N.

ω _i ≤x _i (7)

Mathematical models (3) and (7) represent ω if the ith transition amount prediction timing model is selected _i Greater than 0, if the ith transition amount prediction timing model is not selected, ω _i Equal to 0. The mathematical model (5) indicates that at least 1 flow quantity prediction timing model is selected. The mathematical model (6) indicates that at most P flow quantity prediction timing models are selected. The mathematical model (6) is a constraint that characterizes the number of models to which the constraint is applied.

In practice, the execution body may call an API (Application Programming Interface, application program interface) of a preset solver, and perform solving processing on the objective function to be solved, to obtain a model weight coefficient set. For example, the preset solver may be a Gurobi solver. The preset solver may also be a CPLEX solver. Therefore, the model weight coefficient set obtained by solving processing can be the optimal solution, and each constraint condition in the constraint condition sets can be met.

Optionally, before step 304, the executing body may generate a model error corresponding to the flow amount prediction timing model based on the predicted flow amount group corresponding to each flow amount prediction timing model in the flow amount prediction timing model pool in the predicted flow amount group and each historical flow amount of each return time granularity in the historical flow amount set, so as to obtain a model error set. In practice, the execution entity may generate a model error corresponding to the flow amount prediction timing model using an RMSE (Root Mean Squared Error, root mean square error) formula. The smaller the model error is, the better the corresponding flow quantity prediction time sequence model prediction effect is, and the obtained model error set can be used for adjusting the model weight coefficient in the solving process.

Optionally, after generating the model error corresponding to the flow quantity prediction time sequence model to obtain a model error set, the execution body may delete each constraint condition of the constraint condition set for characterizing and constraining the model weight coefficient, so as to update the constraint condition set. For example, the execution entity may delete mathematical models (1), (2), (3), and (7) and update the constraint set. The updated set of constraints includes mathematical models (4), (5) and (6). Thus, the set of constraints described above may be updated to adapt to the insertion of constraints related to model errors.

Optionally, after deleting each constraint condition in the constraint condition set for characterizing constraint on the model weight coefficient, the executing body may first determine the model error set as input parameters of the first model weight function, the second model weight function, and the third model weight function, so as to generate a first model weight function to be weighted, a second model weight function to be weighted, and a third model weight function to be weighted. The first model weight function, the second model weight function, and the third model weight function may be functions that determine model weight coefficients according to model errors. For example, the first model weight function to be weighted may be:

Wherein, acc _i And representing the model error of the ith circulation quantity prediction time sequence model. e represents the base of the natural logarithmic function.Representing a first model weight function to be weighted.

The second model weight function to be weighted may be:

wherein, the liquid crystal display device comprises a liquid crystal display device,representing a second model weight function to be weighted.

The third model weight function to be weighted may be:

wherein, the liquid crystal display device comprises a liquid crystal display device,representing a third model weight function to be weighted. e represents the base of the natural logarithmic function.

Then, the executing body may perform weighting processing on the first model weight function to be weighted, the second model weight function to be weighted, and the third model weight function to be weighted, so as to obtain a weighted model weight function as a weight constraint condition. For example, the execution body may perform weighting processing on the to-be-weighted first model weight function, the to-be-weighted second model weight function, and the to-be-weighted third model weight function by the following formula to obtain weighted model weight functions:

wherein omega _i Representing a weighted model weight function. h is a ₁ 、h ₂ And h ₃ Satisfy the following requirementsAnd +.>

Finally, the execution body may add the weight constraint to the constraint set to update the constraint set. The constraints included in the updated set of constraints may be expressed as the following mathematical model:

Therefore, the constraint condition set can be updated to insert constraint conditions related to model errors into the constraint condition set, so that the adjustment effect of the model errors on the model weight coefficients can be considered when the model weight coefficients are solved, and the generated model weight coefficient set can improve the prediction accuracy of the multi-time sequence model for the flow quantity prediction.

And 305, carrying out weighted combination processing on the flow quantity prediction time sequence model corresponding to each model weight coefficient in the model weight coefficient set according to the model weight coefficient set to obtain a flow quantity prediction multi-time sequence model.

In some embodiments, the executing body may perform weighted combination processing on the flow amount prediction time sequence model corresponding to each model weight coefficient in the model weight coefficient set according to the model weight coefficient set, to obtain a flow amount prediction multi-time sequence model. The quantity of the circulation quantity prediction time sequence models which are weighted to form the circulation quantity prediction multi-time sequence model meets the constraint condition that the representation is used for constraining the quantity of the models. In practice, first, the execution body may multiply the model weight coefficients with the corresponding flow amount prediction timing models, and then may add the flow amount prediction timing models obtained by multiplying the model weight coefficients to obtain a flow amount prediction multi-timing model. It is to be understood that the above addition may refer to a process of adding the prediction results of the respective flow amount prediction timing models, that is, a process of adding the respective prediction flow amounts. Therefore, the multi-time sequence model for predicting the circulation quantity of the corresponding target object can be obtained by weighting and combining according to the optimal model weight coefficient set.

With further reference to fig. 4, a flow 400 of some embodiments of a method of information transmission is shown. The flow 400 of the information transmission method includes the following steps:

step 401, acquiring a historical circulation quantity set of a target object in a preset time period.

In some embodiments, the execution subject of the information transmission method (for example, the computing device 101 shown in fig. 1 or the computing device 201 shown in fig. 2) may acquire the set of the historical circulation amounts of the target item in the preset period from the terminal storing the historical circulation amounts of the target item through a wired connection manner or a wireless connection manner.

Alternatively, the executing body may acquire the inventory amount of the target item. The stock amount may be a remaining amount of the target item in at least one warehouse storing the target item.

Step 402, inputting the historical circulation quantity set into a circulation quantity prediction multi-time sequence model to obtain a predicted circulation quantity of each prediction time granularity of the target object in a preset prediction time period as a target predicted circulation quantity.

In some embodiments, the execution body may input the historical circulation quantity set into a circulation quantity prediction multi-time sequence model, so as to obtain a predicted circulation quantity of each prediction time granularity of the target object in a preset prediction time period as the target predicted circulation quantity. The flow amount prediction multi-timing model may be the flow amount prediction multi-timing model obtained through steps 301 to 305 in the corresponding embodiments of fig. 3. Therefore, the multi-timing model for predicting the flow quantity can improve the accuracy of predicting the flow quantity of the target. And because the flow quantity prediction multi-timing model is generated for the target object, the influence of the flow quantity change characteristics of objects with different historical time is reduced, and the prediction robustness is improved.

Step 403, transmitting the obtained target predicted stream quantity to the associated display device.

In some embodiments, the executing entity may send the resulting target predicted throughput together to the associated display device. The display device may be a device having a display function and associated with the execution body. Thus, the predicted traffic of each target can be displayed.

Alternatively, the execution body may control the associated scheduling device to execute the scheduling operation according to the inventory amount and the obtained target predicted transfer amount in response to the inventory amount and the obtained target predicted transfer amount satisfying a preset replenishment condition. The preset replenishment condition may be "a difference between a sum of the obtained target predicted transfer amounts and the stock amount is greater than a preset threshold". Here, the specific setting of the preset threshold is not limited. The scheduling device may be a device having a function of scheduling the items. For example, the scheduling device may be an unmanned vehicle. The scheduling operation may be an operation of scheduling an item performed by the scheduling apparatus. In practice, first, the execution subject may determine the difference between the sum of the obtained target predicted traffic amounts and the inventory amount as the scheduling amount. Then, the scheduling device may be controlled to perform scheduling of the scheduling amount of the target items. Therefore, when the current stock quantity is insufficient, the target articles can be scheduled according to the obtained target predicted flow quantity so as to supplement the stock in advance.

The above embodiments of the present disclosure have the following advantageous effects: firstly, a historical circulation quantity set of a target object in a preset time period is obtained. And then, inputting the historical circulation quantity set into a circulation quantity prediction multi-time sequence model to obtain the predicted circulation quantity of each predicted time granularity of the target object in a preset predicted time period as a target predicted circulation quantity. Therefore, the multi-timing model for predicting the flow quantity can improve the accuracy of predicting the flow quantity of the target. And because the flow quantity prediction multi-timing model is generated for the target object, the influence of the flow quantity change characteristics of objects with different historical time is reduced, and the prediction robustness is improved. And finally, the obtained target prediction flow quantity is sent to the associated display equipment. Thus, the predicted traffic of each target can be displayed.

With further reference to fig. 5, as an implementation of the method shown in fig. 3, the present disclosure provides some embodiments of a flow amount prediction multi-time sequence model generating apparatus, which correspond to those method embodiments shown in fig. 3, and which are particularly applicable to various electronic devices.

As shown in fig. 5, the flow amount prediction multi-timing model generation device 500 of some embodiments includes: an acquisition unit 501, an input unit 502, a determination unit 503, a generation unit 504, and a weighted combination unit 505. Wherein the obtaining unit 501 is configured to obtain a set of historical circulation amounts of the target object in a preset time period; the input unit 502 is configured to input the historical circulation quantity set into each circulation quantity prediction time sequence model in a circulation quantity prediction time sequence model pool, and obtain a prediction circulation quantity of each return time granularity of the target object in a preset return time period as a prediction circulation quantity group to form a prediction circulation quantity group set, wherein the preset time period comprises at least one periodic return time period of the preset return time period; the determining unit 503 is configured to determine each historical flow amount of each return time granularity in the set of predicted flow amounts and the set of historical flow amounts as an input parameter of a preset linearization objective function, so as to generate an objective function to be solved; the generating unit 504 is configured to perform solving processing on the objective function to be solved based on a constraint condition set corresponding to the preset linearization objective function, so as to obtain a model weight coefficient set, where a model weight coefficient in the model weight coefficient set corresponds to a flow volume prediction time sequence model in the flow volume prediction time sequence model pool, and the constraint condition set includes a constraint condition for characterizing that the number of models is constrained; the weighted combination unit 505 is configured to perform weighted combination processing on the flow amount prediction time sequence model corresponding to each model weight coefficient in the model weight coefficient set according to the model weight coefficient set, so as to obtain a flow amount prediction multi-time sequence model.

In an alternative implementation of some embodiments, before the generating unit 504, the flow amount prediction multi-timing model generating device 500 may further include: a model error generating unit (not shown in the figure) configured to generate a model error corresponding to the flow amount prediction timing model based on the predicted flow amount group corresponding to each flow amount prediction timing model in the flow amount prediction timing model pool in the predicted flow amount group and each of the historical flow amounts at the respective return time granularities in the historical flow amount set, and obtain a model error set.

In an alternative implementation of some embodiments, after the model error generation unit, the flow amount prediction multi-timing model generation apparatus 500 may further include: a deleting unit (not shown in the figure) configured to delete each constraint condition of the constraint condition set characterizing the constraint on the model weight coefficient to update the constraint condition set.

In an alternative implementation of some embodiments, after deleting the unit, the flow amount prediction multi-timing model generating device 500 may further include: an input parameter determination unit, a weighting processing unit, and an adding unit (not shown in the figure). Wherein the input parameter determination unit is configured to determine the above-mentioned set of model errors as input parameters of the first model weight function, the second model weight function and the third model weight function to generate a first model weight function to be weighted, a second model weight function to be weighted and a third model weight function to be weighted. The weighting processing unit is configured to perform weighting processing on the first model weight function to be weighted, the second model weight function to be weighted and the third model weight function to be weighted, so as to obtain a weighted model weight function as a weight constraint condition. The adding unit is configured to add the above weight constraint to the constraint set to update the constraint set.

It will be appreciated that the elements described in the apparatus 500 correspond to the various steps in the method described with reference to fig. 3. Thus, the operations, features and resulting benefits described above with respect to the method are equally applicable to the apparatus 500 and the units contained therein, and are not described in detail herein.

With further reference to fig. 6, as an implementation of the method shown in fig. 4, the present disclosure provides some embodiments of an information transmission apparatus, which correspond to those method embodiments shown in fig. 4, and which are particularly applicable in various electronic devices.

As shown in fig. 6, the information transmission apparatus 600 of some embodiments includes: a history flow amount set acquisition unit 601, a history flow amount set input unit 602, and a transmission unit 603. The historical circulation quantity set obtaining unit 601 is configured to obtain a historical circulation quantity set of a target object in a preset time period; the historical circulation quantity set input unit 602 is configured to input the historical circulation quantity set into a circulation quantity prediction multi-time sequence model, so as to obtain a predicted circulation quantity of each prediction time granularity of the target object in a preset prediction time period as a target predicted circulation quantity; the transmitting unit 603 is configured to transmit the resulting target predicted stream quantity to an associated display device.

In an alternative implementation of some embodiments, the historical flow amount set obtaining unit 601 may further include: an inventory amount acquisition unit (not shown in the figure) configured to acquire an inventory amount of the above-described target article.

In an alternative implementation of some embodiments, the information sending apparatus 600 may further include: a control unit (not shown in the figure) configured to control the associated scheduling device to perform a scheduling operation according to the above-described stock quantity and the obtained target predicted transfer quantity in response to the above-described stock quantity and the obtained target predicted transfer quantity satisfying a preset replenishment condition.

It will be appreciated that the elements described in the apparatus 600 correspond to the various steps in the method described with reference to fig. 4. Thus, the operations, features and resulting benefits described above with respect to the method are equally applicable to the apparatus 600 and the units contained therein, and are not described in detail herein.

Referring now to FIG. 7, a schematic diagram of a structure of an electronic device (e.g., computing device 101 of FIG. 1 or computing device 201 of FIG. 2) 700 suitable for use in implementing some embodiments of the disclosure is shown. The electronic device shown in fig. 7 is only one example and should not impose any limitations on the functionality and scope of use of embodiments of the present disclosure.

As shown in fig. 7, the electronic device 700 may include a processing means (e.g., a central processor, a graphics processor, etc.) 701, which may perform various appropriate actions and processes according to a program stored in a Read Only Memory (ROM) 702 or a program loaded from a storage means 708 into a Random Access Memory (RAM) 703. In the RAM703, various programs and data required for the operation of the electronic device 700 are also stored. The processing device 701, the ROM 702, and the RAM703 are connected to each other through a bus 704. An input/output (I/O) interface 705 is also connected to bus 704.

In general, the following devices may be connected to the I/O interface 705: input devices 706 including, for example, a touch screen, touchpad, keyboard, mouse, camera, microphone, accelerometer, gyroscope, and the like; an output device 707 including, for example, a Liquid Crystal Display (LCD), a speaker, a vibrator, and the like; storage 708 including, for example, magnetic tape, hard disk, etc.; and a communication device 709. The communication means 709 may allow the electronic device 700 to communicate wirelessly or by wire with other devices to exchange data. While fig. 7 shows an electronic device 700 having various means, it is to be understood that not all of the illustrated means are required to be implemented or provided. More or fewer devices may be implemented or provided instead. Each block shown in fig. 7 may represent one device or a plurality of devices as needed.

In particular, according to some embodiments of the present disclosure, the processes described above with reference to flowcharts may be implemented as computer software programs. For example, some embodiments of the present disclosure include a computer program product comprising a computer program embodied on a computer readable medium, the computer program comprising program code for performing the method shown in the flow chart. In such embodiments, the computer program may be downloaded and installed from a network via communications device 709, or from storage 708, or from ROM 702. The above-described functions defined in the methods of some embodiments of the present disclosure are performed when the computer program is executed by the processing means 701.

It should be noted that, the computer readable medium described in some embodiments of the present disclosure may be a computer readable signal medium or a computer readable storage medium, or any combination of the two. The computer readable storage medium can be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or a combination of any of the foregoing. More specific examples of the computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer diskette, 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 disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In some embodiments of the present disclosure, a computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. In some embodiments of the present disclosure, however, the computer-readable signal medium may comprise a data signal propagated in baseband or as part of a carrier wave, with the computer-readable program code embodied therein. Such a propagated data signal may take any of a variety of forms, including, but not limited to, electro-magnetic, optical, or any suitable combination of the foregoing. A computer readable signal medium may also be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to: electrical wires, fiber optic cables, RF (radio frequency), and the like, or any suitable combination of the foregoing.

In some implementations, the clients, servers may communicate using any currently known or future developed network protocol, such as HTTP (HyperText Transfer Protocol ), and may be interconnected with any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network ("LAN"), a wide area network ("WAN"), the internet (e.g., the internet), and peer-to-peer networks (e.g., ad hoc peer-to-peer networks), as well as any currently known or future developed networks.

The computer readable medium may be contained in the electronic device; or may exist alone without being incorporated into the electronic device. The computer readable medium carries one or more programs which, when executed by the electronic device, cause the electronic device to: acquiring a historical circulation quantity set of a target object in a preset time period; inputting the historical circulation quantity set into each circulation quantity prediction time sequence model in a circulation quantity prediction time sequence model pool, and obtaining a prediction circulation quantity of each return time granularity of the target object in a preset return time period as a prediction circulation quantity group to form a prediction circulation quantity group set, wherein the preset time period comprises at least one periodic return time period of the preset return time period; determining each historical circulation quantity of each return time granularity in the predicted circulation quantity group set and the historical circulation quantity set as an input parameter of a preset linearization objective function so as to generate an objective function to be solved; solving the objective function to be solved based on a constraint condition set corresponding to the preset linearization objective function to obtain a model weight coefficient set, wherein the model weight coefficient in the model weight coefficient set corresponds to the flow quantity prediction time sequence model in the flow quantity prediction time sequence model pool, and the constraint condition set comprises constraint conditions for representing constraint on the number of models; and carrying out weighted combination processing on the circulation quantity prediction time sequence model corresponding to each model weight coefficient in the model weight coefficient set according to the model weight coefficient set to obtain a circulation quantity prediction multi-time sequence model.

Computer program code for carrying out operations for some embodiments of the present disclosure may be written in one or more programming languages, including an object oriented programming language such as Java, smalltalk, C ++ and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any kind of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or may be connected to an external computer (for example, through the Internet using an Internet service provider).

The flowcharts and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present disclosure. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.

The units described in some embodiments of the present disclosure may be implemented by means of software, or may be implemented by means of hardware. The described units may also be provided in a processor, for example, described as: a processor includes an acquisition unit, an input unit, a determination unit, a generation unit, and a weighted combination unit. The names of these units do not constitute a limitation on the unit itself in some cases, and for example, the acquisition unit may also be described as "a unit that acquires a set of historical circulation amounts of the target article in a preset period of time".

The functions described above herein may be performed, at least in part, by one or more hardware logic components. For example, without limitation, exemplary types of hardware logic components that may be used include: a Field Programmable Gate Array (FPGA), an Application Specific Integrated Circuit (ASIC), an Application Specific Standard Product (ASSP), a system on a chip (SOC), a Complex Programmable Logic Device (CPLD), and the like.

The foregoing description is only of the preferred embodiments of the present disclosure and description of the principles of the technology being employed. It will be appreciated by those skilled in the art that the scope of the invention in the embodiments of the present disclosure is not limited to the specific combination of the above technical features, but encompasses other technical features formed by any combination of the above technical features or their equivalents without departing from the spirit of the invention. Such as the above-described features, are mutually substituted with (but not limited to) the features having similar functions disclosed in the embodiments of the present disclosure.

Claims

1. A method for generating a multi-time sequence model of flow quantity prediction comprises the following steps:

acquiring a historical circulation quantity set of a target object in a preset time period;

inputting the historical circulation quantity set into each circulation quantity prediction time sequence model in a circulation quantity prediction time sequence model pool, and obtaining a prediction circulation quantity of each return time granularity of the target object in a preset return time period as a prediction circulation quantity group to form a prediction circulation quantity group set, wherein the preset time period comprises at least one periodic return time period of the preset return time period;

determining each historical circulation quantity of each return time granularity in the predicted circulation quantity group set and the historical circulation quantity set as an input parameter of a preset linearization objective function so as to generate an objective function to be solved;

solving the objective function to be solved based on a constraint condition set corresponding to the preset linearization objective function to obtain a model weight coefficient set, wherein the model weight coefficient in the model weight coefficient set corresponds to a flow quantity prediction time sequence model in the flow quantity prediction time sequence model pool, and the constraint condition set comprises constraint conditions for representing constraint on the number of models;

And carrying out weighted combination treatment on the circulation quantity prediction time sequence model corresponding to each model weight coefficient in the model weight coefficient set according to the model weight coefficient set to obtain a circulation quantity prediction multi-time sequence model.

2. The method of claim 1, wherein prior to the solving the objective function to be solved, the method further comprises:

generating model errors corresponding to the circulation quantity prediction time sequence model based on the prediction circulation quantity group corresponding to each circulation quantity prediction time sequence model in the circulation quantity prediction time sequence model pool and each historical circulation quantity in each return time granularity in the historical circulation quantity set, and obtaining a model error set.

3. The method of claim 2, wherein after said generating model errors corresponding to the flow quantity prediction timing model, resulting in a set of model errors, the method further comprises:

and deleting each constraint condition of the constraint condition set characterization, which is used for constraining the model weight coefficient, so as to update the constraint condition set.

4. A method according to claim 3, wherein after said deleting the respective constraints in the set of constraints that characterize the constraints on model weight coefficients, the method further comprises:

Determining the model error set as input parameters of a first model weight function, a second model weight function and a third model weight function to generate a first model weight function to be weighted, a second model weight function to be weighted and a third model weight function to be weighted;

weighting the first model weight function to be weighted, the second model weight function to be weighted and the third model weight function to be weighted to obtain a weighted model weight function as a weight constraint condition;

the weight constraint is added to the constraint set to update the constraint set.

5. An information transmission method, comprising:

inputting the historical circulation quantity set into a circulation quantity prediction multi-time sequence model to obtain a prediction circulation quantity of each prediction time granularity of the target object in a preset prediction time period as a target prediction circulation quantity, wherein the circulation quantity prediction multi-time sequence model is generated by adopting the method as set forth in any one of claims 1-4;

and sending the obtained target predicted stream quantity to an associated display device.

6. The method of claim 5, wherein the obtaining a set of historical circulation amounts for the target item over a preset period of time further comprises:

And acquiring the stock quantity of the target object.

7. The method of claim 6, wherein the method further comprises:

and controlling associated dispatching equipment to execute dispatching operation according to the stock quantity and the obtained target predicted flow quantity in response to the stock quantity and the obtained target predicted flow quantity meeting preset replenishment conditions.

8. A flow quantity prediction multi-timing model generation device comprises:

an acquisition unit configured to acquire a set of historical circulation amounts of a target article within a preset time period;

the input unit is configured to input the historical circulation quantity set into each circulation quantity prediction time sequence model in a circulation quantity prediction time sequence model pool, and obtain a prediction circulation quantity of each return time granularity of the target object in a preset return time period as a prediction circulation quantity group to form a prediction circulation quantity group set, wherein the preset time period comprises at least one periodic return time period of the preset return time period;

a determining unit configured to determine each historical circulation quantity of each return time granularity in the predicted circulation quantity group set and the historical circulation quantity set as an input parameter of a preset linearization objective function so as to generate an objective function to be solved;

The generation unit is configured to carry out solving processing on the objective function to be solved based on a constraint condition set corresponding to the preset linearization objective function to obtain a model weight coefficient set, wherein the model weight coefficient in the model weight coefficient set corresponds to the flow quantity prediction time sequence model in the flow quantity prediction time sequence model pool, and the constraint condition set comprises constraint conditions for representing constraint on the number of models;

and the weighted combination unit is configured to perform weighted combination processing on the circulation quantity prediction time sequence model corresponding to each model weight coefficient in the model weight coefficient set according to the model weight coefficient set to obtain a circulation quantity prediction multi-time sequence model.

9. An information transmitting apparatus comprising:

a history flow amount set obtaining unit configured to obtain a history flow amount set of a target article in a preset time period;

a historical circulation quantity set input unit configured to input the historical circulation quantity set into a circulation quantity prediction multi-time sequence model, so as to obtain a predicted circulation quantity of each predicted time granularity of the target object in a preset predicted time period as a target predicted circulation quantity, wherein the circulation quantity prediction multi-time sequence model is generated by adopting the method as set forth in any one of claims 1-4;

And a transmitting unit configured to transmit the obtained target predicted stream quantity to an associated display device.

10. An electronic device, comprising:

one or more processors;

a storage device having one or more programs stored thereon,

when executed by the one or more processors, causes the one or more processors to implement the method of any of claims 1-7.

11. A computer readable medium having stored thereon a computer program, wherein the program when executed by a processor implements the method of any of claims 1-7.