CN110533437B - Advertisement delivery budget allocation method and device - Google Patents
Advertisement delivery budget allocation method and device Download PDFInfo
- Publication number
- CN110533437B CN110533437B CN201810502118.2A CN201810502118A CN110533437B CN 110533437 B CN110533437 B CN 110533437B CN 201810502118 A CN201810502118 A CN 201810502118A CN 110533437 B CN110533437 B CN 110533437B
- Authority
- CN
- China
- Prior art keywords
- budget allocation
- index
- delivery channel
- budget
- effect
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0241—Advertisements
- G06Q30/0247—Calculate past, present or future revenues
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0241—Advertisements
- G06Q30/0249—Advertisements based upon budgets or funds
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Accounting & Taxation (AREA)
- Development Economics (AREA)
- Strategic Management (AREA)
- Finance (AREA)
- Game Theory and Decision Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Economics (AREA)
- Marketing (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a method and a device for distributing budget of advertisement delivery, wherein the method establishes an index prediction model through fitting prediction to obtain effect index predicted values corresponding to various budget distribution combinations of each delivery channel; and optimizing and solving the effect index predicted values corresponding to various budget allocation combinations by using a genetic algorithm to obtain a plurality of budget allocation combinations with optimal effect indexes, and performing budget allocation by using the method to maximize the profit of the advertisement publisher.
Description
Technical Field
The invention relates to the technical field of computers, in particular to an advertisement delivery budget allocation method and device.
Background
Currently, in the internet age, internet advertising has become the mainstream. The internet advertisement has the advantages of wide coverage, strong initiative and enthusiasm, relatively low cost, high cost performance and the like.
A plurality of delivery channels and delivery devices are involved in Internet advertisement delivery, even in the same media device classification, a plurality of accounts, plans, units and the like exist, and advertisement delivery is more finely managed. In the release management process, the optimized combination of a plurality of service indexes is involved. For example, given a consumption budget, how to consume can minimize the overall Charge Per Click (CPC) and maximize the Click-through rate CTR. Traditional advertisement delivery management methods have not given a solution to how to allocate an advertisement delivery budget to maximize multiple benefit objectives.
Disclosure of Invention
In view of the foregoing problems, the present invention provides a method and an apparatus for allocating an advertisement delivery budget to solve the technical problem that the traditional advertisement delivery management scheme cannot achieve maximization of multiple benefit targets.
In a first aspect, the present application provides an advertisement placement budget allocation method, including:
performing fitting prediction according to historical release data and effect data corresponding to each release channel respectively to obtain at least one index prediction model corresponding to each release channel;
obtaining at least one budget allocation combination corresponding to each delivery channel according to the budget total, wherein each budget allocation combination comprises the budget allocation amount corresponding to each delivery channel;
for any budget allocation combination, respectively predicting each effect index predicted value corresponding to the current budget allocation amount of the corresponding delivery channel by using an index prediction model corresponding to each delivery channel, wherein the effect index is an index for representing the advertisement delivery effect;
optimizing and solving effect index predicted values corresponding to all budget allocation combinations by using a genetic algorithm to obtain a non-inferior solution set, wherein each solution in the non-inferior solution set is a budget allocation combination corresponding to each delivery channel;
and determining a target budget allocation combination from the non-inferior solution set.
Optionally, the optimizing and solving are performed on the effect index predicted values corresponding to all budget allocation combinations by using a genetic algorithm to obtain a non-inferior solution set, including:
setting an optimization target according to the service requirement, and setting a variation probability;
selecting at least two gene sequences according to the optimization target, and performing gene evolution operation on the selected gene sequences according to the variation probability to generate a next generation gene sequence;
carrying out the gene evolution operation on each generation of gene sequence until the iteration times reach the specified iteration times to obtain the non-inferior solution set;
and the corresponding gene segments of all the delivery channels form a gene sequence.
Optionally, the fitting prediction is performed on historical delivery data and effect data corresponding to each delivery channel, so as to obtain at least one index prediction model corresponding to each delivery channel, and the method includes:
the following steps are performed for one delivery channel:
acquiring a consumption amount and effect data corresponding to any effect index from historical release data corresponding to the release channel;
and according to the consumption amount and the effect data corresponding to the effect index, performing fitting prediction to obtain an index prediction model corresponding to the effect index.
Optionally, determining a target budget allocation combination from the non-inferior solution set includes:
selecting a solution meeting the overall constraint condition of each effect index from the non-inferior solution set to obtain a target budget allocation combination;
alternatively, the first and second electrodes may be,
and selecting the optimal solution of each effect index from the non-inferior solution set to obtain a target budget allocation combination.
Optionally, the solutions in the non-inferior solution set are the remaining solutions after the solution with the worst effect index is removed.
Optionally, the obtaining at least one budget allocation combination corresponding to each delivery channel according to the budget total includes:
and obtaining at least one budget allocation combination corresponding to each delivery channel according to the budget total and the constraint conditions of each delivery channel.
Optionally, the historical delivery data includes delivery channel categories and budget allocations; the effect data comprises advertisement display amount, click rate and conversion amount.
In a second aspect, the present application provides an advertisement placement budget allocation method, including:
the fitting prediction module is used for performing fitting prediction according to historical delivery data and effect data corresponding to each delivery channel respectively to obtain at least one index prediction model corresponding to each delivery channel;
the budget allocation module is used for acquiring at least one budget allocation combination corresponding to each delivery channel according to the budget total amount, and each budget allocation combination comprises the budget allocation amount corresponding to each delivery channel;
the forecasting module is used for forecasting each effect index forecasting value corresponding to the current budget allocation amount of the corresponding delivery channel by using the index forecasting model corresponding to each delivery channel aiming at any budget allocation combination, and the effect index is an index for representing the advertisement delivery effect;
the optimization module is used for carrying out optimization solution on the effect index predicted values corresponding to all the budget allocation combinations by utilizing a genetic algorithm to obtain a non-inferior solution set, wherein each solution in the non-inferior solution set is a budget allocation combination corresponding to each delivery channel;
and the determining module is used for determining the target budget allocation combination from the non-inferior solution set.
In a third aspect, the present application provides a storage medium, on which a program is stored, where the program, when executed by a processor, implements the method for allocating an advertisement delivery budget according to any one of the possible implementation manners of the first aspect.
In a fourth aspect, the present application provides a processor, where the processor is configured to execute a program, and the program executes the method for allocating an advertisement delivery budget according to any one of the possible implementation manners of the first aspect.
The advertisement delivery budget allocation method provided in this embodiment performs fitting prediction by using historical delivery data and effect data corresponding to a delivery channel to obtain at least one index prediction model corresponding to the delivery channel. Acquiring at least one budget allocation combination corresponding to each delivery channel; and aiming at each budget allocation combination, predicting the effect index predicted value corresponding to the current budget allocation amount of the corresponding delivery channel by using the index prediction model corresponding to each delivery channel. And optimizing and solving the effect index predicted values corresponding to all budget allocation combinations by using a genetic algorithm to obtain a non-inferior solution set. And finally, selecting the budget allocation combination with the optimal effect index from the non-inferior solution set. The method comprises the steps of establishing an index prediction model through fitting prediction to obtain effect index prediction values corresponding to various budget allocation combinations of various delivery channels; and optimizing and solving the effect index predicted values corresponding to various budget allocation combinations by using a genetic algorithm to obtain a plurality of budget allocation combinations with optimal effect indexes, and performing budget allocation by using the method to maximize the profit of the advertisement publisher.
The foregoing description is only an overview of the technical solutions of the present invention, and the embodiments of the present invention are described below in order to make the technical means of the present invention more clearly understood and to make the above and other objects, features, and advantages of the present invention more clearly understandable.
Drawings
Various additional advantages and benefits will become apparent to those of ordinary skill in the art upon reading the following detailed description of the preferred embodiments. The drawings are only for purposes of illustrating the preferred embodiments and are not to be construed as limiting the invention. Also, like reference numerals are used to refer to like parts throughout the drawings. In the drawings:
fig. 1 is a schematic flowchart illustrating an advertisement placement budget allocation method according to an embodiment of the present application;
FIG. 2 is a flow chart illustrating a process of creating an index prediction model according to an embodiment of the present disclosure;
FIG. 3 is a flowchart illustrating another method for allocating budget for advertisement placement according to an embodiment of the present disclosure;
fig. 4 is a block diagram illustrating an apparatus for allocating an advertisement delivery budget according to an embodiment of the present application;
FIG. 5 is a block diagram of an optimization module according to an embodiment of the present application.
Detailed Description
Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. While exemplary 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 limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.
Referring to fig. 1, a flowchart of an advertisement placement budget allocation method for allocating an advertisement placement budget of an advertisement publisher to maximize the benefit of the advertisement publisher according to an embodiment of the present application is shown. The method is applied to a terminal or a server, and as shown in fig. 1, the method provided by this embodiment may include:
and S110, performing fitting prediction according to the historical delivery data and the effect data corresponding to each delivery channel to obtain at least one index prediction model corresponding to each delivery channel.
The delivery channels refer to different platforms for delivering advertisements, for example, including search platforms such as hundred-degree PC, hundred-degree mobile, 360PC, 360mobile, and the like, social platforms such as WeChat, microblog, and the like, and E-commerce platforms such as Taobao, Jingdong, and the like.
And acquiring historical delivery data and corresponding effect data of each delivery channel. The historical placement data is placement data for a placement channel over a past period of time, for example, the historical placement data may include a consumption amount for the placement channel. The effectiveness data is data representing the effectiveness of advertisement delivery of the delivery channel, and for example, the effectiveness data mainly includes various index data such as advertisement display amount, click rate or conversion amount.
In an embodiment of the application, for one delivery channel, a plurality of index prediction models, for example, a prediction model of the consumption amount and the click rate and a prediction model between the consumption amount and the exposure amount, can be established according to business requirements.
S111 to S112 shown in fig. 2 are executed for any delivery channel:
and S111, acquiring the consumption amount and the effect data corresponding to any effect index from the historical delivery data corresponding to the delivery channel.
And S112, performing fitting prediction according to the consumption amount and the effect data corresponding to the effect index to obtain an index prediction model corresponding to the effect index.
In one embodiment of the present application, the fitting method employed for fitting the prediction includes, but is not limited to: linear regression, polynomial regression, logistic regression, random forest algorithm, gradient boosting decision tree (GBRT) algorithm, and the like.
And S120, obtaining at least one budget allocation combination corresponding to each delivery channel according to the budget total, wherein each budget allocation combination comprises the budget allocation amount corresponding to each delivery channel.
And distributing the total budget into each delivery channel according to the historical consumption proportion of each channel within a historical period of time and floating a certain range to obtain a plurality of budget distribution combinations, namely, obtaining a plurality of combination modes of budget distribution corresponding to each delivery channel.
In another embodiment of the present application, in the budget allocation process, budget allocation is performed according to the constraint conditions of each delivery channel, so as to obtain a plurality of budget allocation combinations. For example, assuming that the delivery channel is two channels a and B and the budget total is 6000 yuan, different budget quotas can be allocated to the channel a and the channel B within the budget of 6000 yuan. For example, 4000 yuan for channel a and 2000 yuan for channel B may be allocated as a budget allocation combination; while allocating 1000 for channel a and 5000 for channel B may be another budget allocation combination. Within the constraint condition range of the channel A and the channel B, various budget allocation combinations can be obtained.
The constraint conditions of the delivery channels are set according to the requirements of the advertisement deliverers, for example, the budget amount of a mobile terminal is required to be higher than that of a PC terminal by a certain advertisement deliverer, and the delivery budget of each delivery channel floats within 20% of the delivery ratio of each historical channel.
And S130, for each budget allocation combination, respectively predicting each effect index predicted value corresponding to the current budget allocation amount of the corresponding delivery channel by using the index prediction model corresponding to each delivery channel.
The effectiveness index is used for representing the effectiveness of the advertisement, such as exposure, click rate, and the like.
And for any budget allocation combination obtained in the step S120, predicting the predicted value of each effect index corresponding to the current budget allocation amount of the delivery channel by using the index prediction model corresponding to the corresponding delivery channel.
For example, assume that there are two delivery channels a and B, where one budget allocation combination is: the budget allocation of the channel A is 1000, and the budget allocation of the channel B is 5000. And predicting what the click rate is brought when the consumption amount of the channel A is 1000 yuan by using the predictive models of the consumption amount of the channel A and the click rate. And predicting the exposure amount brought when the consumption amount of the channel A is 1000 yuan by using a prediction model of the consumption amount and the exposure amount of the channel A. And similarly, predicting the predicted value of each effect index of the channel B by using the index prediction model.
And S140, optimizing and solving the effect index predicted values corresponding to all budget allocation combinations by using a genetic algorithm to obtain a non-inferior solution set.
Each solution in the non-inferior solution set is a budget allocation combination corresponding to each delivery channel.
In one embodiment of the present application, the solutions in the non-inferior solution set are the remaining solutions after removing the solution with the worst effect index.
The genetic algorithm is a calculation model of a biological evolution process for simulating natural selection and genetic mechanism of Darwinian biological evolution theory, and is a method for searching an optimal solution by simulating the natural evolution process. The genetic algorithm simulates the problem to be solved into a biological evolution process, generates the next generation solution through operations such as copying, crossing, mutation and the like, gradually eliminates the solution with low fitness function value, and increases the solution with high fitness function value. Thus, a solution with a high fitness function value is likely to be evolved after N generations of evolution.
In the method, each delivery channel is used as a gene block in the process of solving by using a genetic algorithm, a budget allocation amount and a corresponding effect index predicted value of the delivery channel are gene segments in the gene block, and any gene segment of all the delivery channels forms a gene sequence. Some gene sequences are randomly selected, and the gene sequences of the next generation are generated through operations of crossing, mutation and the like. Wherein, the probability that the gene segment with better effect index prediction value is selected is higher. Thus, based on the set optimization target, a non-inferior solution set corresponding to a plurality of index combinations may be generated after the evolution of the G generation. The optimization target may be set according to specific business requirements, for example, the click rate is as large as possible, and the consumption amount is as small as possible.
And S150, determining a target budget allocation combination from the non-inferior solution set.
In one embodiment of the present application, the advertiser has a certain requirement on the performance index, for example, a certain client requires that the average CPC must be less than 1.5 yuan, on the basis of which the CTR is as large as possible. Under the application scene, according to the overall constraint condition of each effect index, a solution meeting the requirement is selected from the non-inferior solution set, namely, the target budget allocation combination.
In another embodiment of the present application, a solution with a better effect index can be directly selected as a final target budget allocation combination.
In the advertisement delivery budget allocation method provided by this embodiment, an index prediction model corresponding to each delivery channel is established through fitting prediction, and an effect index prediction value corresponding to each budget allocation combination of each channel is obtained; and performing optimization solution on the effect index predicted values corresponding to various budget allocation combinations to obtain a plurality of budget allocation combinations with optimal effect indexes, and performing budget allocation by using the method to maximize the profit of the advertisement publisher.
Referring to fig. 3, a flowchart of another method for allocating an advertisement delivery budget according to an embodiment of the present application is shown, and this embodiment will focus on a process of performing an optimization solution by using a genetic algorithm, as shown in fig. 3, the method may include:
s210, acquiring historical data corresponding to each delivery channel. The historical data includes historical delivery data and corresponding effect data.
And S220, performing fitting prediction according to the historical delivery data and the effect data corresponding to each delivery channel to obtain at least one index prediction model corresponding to each delivery channel.
In an embodiment of the application, for one delivery channel, a plurality of index prediction models, for example, a prediction model of the consumption amount and the click rate and a prediction model between the consumption amount and the exposure amount, can be established according to business requirements.
And S230, obtaining at least one budget allocation combination corresponding to each delivery channel according to the budget total.
Each budget allocation combination comprises budget allocations corresponding to the delivery channels.
S240, aiming at any budget allocation combination, predicting each effect index predicted value corresponding to the current budget allocation amount of the delivery channel by using the index prediction model corresponding to each delivery channel.
S250, setting an optimization target and a mutation probability.
An optimization target can be set according to the service requirement of advertisement delivery, for example, the optimization target is within a certain budget range, less budget is spent as much as possible, and the click rate is maximized; the click rate is a composite index, so the click rate needs to be disassembled into basic indexes in a specific optimization process, and the final optimization target is as follows: the click quantity is as large as possible, the exposure quantity is as small as possible, and the budget is as small as possible.
The mutation probability is set according to actual requirements, for example, the mutation probability may be set to 0.2-0.3.
S260, selecting a gene sequence according to the optimization target, and performing gene evolution operation on the selected gene sequence according to the mutation probability to generate a gene sequence of the next generation.
The genetic evolution operation comprises operations such as replication, crossing, mutation, selection and the like.
In the optimization solving process by using the genetic algorithm, at least two gene sequences are selected according to an optimization target, and then gene evolution operation is performed on the selected at least two gene sequences according to the variation probability to obtain the gene sequences of the next generation corresponding to the at least two gene sequences.
S270, carrying out the gene evolution operation aiming at the gene sequence of each generation to obtain a non-inferior solution set.
And performing S260 operation on each obtained gene sequence generation until the iteration number reaches the specified iteration number, and stopping iteration. The specified iteration number can be set according to specific service requirements and time requirements.
In an embodiment of the application, a non-inferior solution set is screened out by a solution obtained by iterative computation of a gene sequence, and the gene sequence obtained in the iterative computation process needs to be substituted into an index prediction model of a corresponding delivery channel for prediction to obtain an effect index prediction value corresponding to the budget allocation combination. And then, removing the solutions with poor effect indexes according to the effect index predicted values to obtain a non-inferior solution set. In the method, the corresponding effect index is calculated only by substituting the solution obtained after the iterative process is finished into the index prediction model, so that the time consumed in the whole optimization process is short.
In another embodiment of the present application, in each iterative calculation process of the gene sequence, the obtained gene sequence is substituted into the corresponding index prediction model to predict the corresponding effect index prediction value. Then, all the effect indexes are removed, and the next iteration process is carried out. In this way, each iteration needs to substitute the obtained solution into the index prediction model to calculate the corresponding effect index, so that the whole optimization process takes long time. S280, determining a target budget allocation combination from the non-inferior solution set.
And finally, determining a solution which meets the optimization target, namely the target budget allocation combination, from the obtained non-inferior solution set.
In the method for allocating budget for advertisement delivery provided in this embodiment, an index prediction model is established by using a fitting algorithm, and the effect index prediction values corresponding to various budget allocation combinations of each delivery channel are predicted by using the index prediction model. The effect index predicted values corresponding to various budget allocation combinations are optimized and solved by using a genetic algorithm to obtain a plurality of budget allocation combinations with optimal effect indexes, and the method is used for budget allocation to maximize the profit of an advertisement publisher.
Corresponding to the embodiment of the method for allocating the advertisement delivery budget, the application also provides an embodiment of a device for allocating the advertisement delivery budget.
Referring to fig. 4, a block diagram of an advertisement delivery budget allocation apparatus applied in a terminal or a server according to an embodiment of the present application is shown. The apparatus may include: a fitting prediction module 110, a budget allocation module 120, a prediction module 130, an optimization module 140, and a determination module 150;
and the fitting prediction module 110 is configured to perform fitting prediction according to the historical delivery data and the effect data corresponding to each delivery channel, respectively, to obtain at least one index prediction model corresponding to each delivery channel.
The historical delivery data comprises delivery channel categories and budget allocation; the effect data comprises advertisement display amount, click rate and conversion amount.
For a delivery channel, the fitting prediction module 110 is specifically configured to: acquiring a consumption amount and effect data corresponding to any effect index from historical release data corresponding to a release channel; and according to the consumption amount and the effect data corresponding to the effect index, performing fitting prediction to obtain an index prediction model corresponding to the effect index.
The budget allocation module 120 is configured to obtain at least one budget allocation combination corresponding to each delivery channel according to the budget total, where each budget allocation combination includes a budget allocation amount corresponding to each delivery channel.
In some application scenes, an advertisement publisher has certain constraint limits on each delivery channel, and in the application scenes, at least one budget allocation combination corresponding to each delivery channel is obtained according to the budget total and the constraint conditions of each delivery channel.
And the predicting module 130 is configured to predict, for any budget allocation combination, each effect index prediction value corresponding to the current budget allocation amount of the corresponding delivery channel by using the index prediction model corresponding to each delivery channel.
The effect index is an index for representing the advertisement putting effect, such as exposure, click rate, and the like;
and the optimization module 140 is configured to perform optimization solution on the effect index predicted values corresponding to all budget allocation combinations by using a genetic algorithm to obtain a non-inferior solution set.
Each solution in the non-inferior solution set is a budget allocation combination corresponding to each delivery channel.
Referring to fig. 5, a block diagram of an optimization module according to an embodiment of the present application is shown, where the optimization module 140 may include:
the setting submodule 141 is configured to set an optimization target according to a service requirement, and set a mutation probability.
And an evolution submodule 142, configured to select at least two gene sequences according to the optimization target, and perform a gene evolution operation on the selected gene sequences according to the variation probability to generate a next-generation gene sequence.
And the iteration submodule 143 is configured to perform a gene evolution operation on each generation of gene sequence until the iteration number reaches a specified iteration number, so as to obtain the non-inferior solution set.
And the gene segment corresponding to the delivery channel forms a gene sequence.
A determining module 150, configured to determine a target budget allocation combination from the non-inferior solution set.
In some application scenarios, constraint conditions are set for the effect indexes, and in such application scenarios, solutions meeting the overall constraint conditions of each effect index are selected from the non-inferior solution set to obtain a target budget allocation combination.
In another application scenario of the present application, a solution with an optimal effect index is selected from a non-inferior solution set as a target budget allocation combination.
The advertisement delivery budget allocation device provided in this embodiment establishes an index prediction model corresponding to each delivery channel through fitting prediction, and obtains an effect index prediction value corresponding to each budget allocation combination of each channel; and performing optimization solution on the effect index predicted values corresponding to various budget allocation combinations to obtain a plurality of budget allocation combinations with optimal effect indexes, and performing budget allocation by using the method to maximize the profit of the advertisement publisher.
The advertisement delivery budget allocation device comprises a processor and a memory, wherein the fitting prediction module 110, the budget allocation module 120, the prediction module 130, the optimization module 140, the determination module 150 and the like are stored in the memory as program units, and the processor executes the program units stored in the memory to realize corresponding functions.
The processor comprises a kernel, and the kernel calls the corresponding program unit from the memory. The kernel can be set to be one or more than one, and advertisement delivery budget allocation is carried out by adjusting kernel parameters, so that the benefit of an advertisement delivery person is maximized.
The memory may include volatile memory in a computer readable medium, Random Access Memory (RAM) and/or nonvolatile memory such as Read Only Memory (ROM) or flash memory (flash RAM), and the memory includes at least one memory chip.
An embodiment of the present invention provides a storage medium, on which a program is stored, where the program, when executed by a processor, implements any of the above embodiments of the advertisement delivery budget allocation method.
The embodiment of the invention provides a processor, which is used for running a program, wherein the program runs to execute any one of the embodiments of the advertisement delivery budget allocation method.
The embodiment of the invention provides equipment, which comprises a processor, a memory and a program which is stored on the memory and can run on the processor, wherein the processor executes the program and realizes the following steps:
performing fitting prediction according to historical delivery data and effect data corresponding to each delivery channel respectively to obtain at least one index prediction model corresponding to each delivery channel;
obtaining at least one budget allocation combination corresponding to each delivery channel according to the budget total, wherein each budget allocation combination comprises the budget allocation amount corresponding to each delivery channel;
for any budget allocation combination, respectively predicting each effect index predicted value corresponding to the current budget allocation amount of the corresponding delivery channel by using an index prediction model corresponding to each delivery channel, wherein the effect index is an index for representing the advertisement delivery effect;
optimizing and solving effect index predicted values corresponding to all budget allocation combinations by using a genetic algorithm to obtain a non-inferior solution set, wherein each solution in the non-inferior solution set is a budget allocation combination corresponding to each delivery channel;
and determining a target budget allocation combination from the non-inferior solution set.
In one possible implementation manner of the present application, the optimizing and solving the effect index predicted values corresponding to all budget allocation combinations by using a genetic algorithm to obtain a non-inferior solution set includes:
setting an optimization target according to the service requirement, and setting a variation probability;
selecting at least two gene sequences according to the optimization target, and performing gene evolution operation on the selected gene sequences according to the variation probability to generate a next generation gene sequence;
carrying out the gene evolution operation on each generation of gene sequence until the iteration times reach the specified iteration times to obtain the non-inferior solution set;
and the corresponding effect index predicted value and any budget allocation amount of any delivery channel form a gene segment, and the gene segments corresponding to all the delivery channels form a gene sequence.
In another possible implementation manner of the present application, the fitting prediction is performed on historical delivery data and effect data corresponding to each delivery channel, so as to obtain at least one index prediction model corresponding to each delivery channel, including:
the following steps are performed for one delivery channel:
acquiring a consumption amount and effect data corresponding to any effect index from historical release data corresponding to the release channel;
and according to the consumption amount and the effect data corresponding to the effect index, performing fitting prediction to obtain an index prediction model corresponding to the effect index.
In another possible implementation manner of the present application, determining a target budget allocation combination from the non-inferior solution set includes:
selecting a solution meeting the overall constraint condition of each effect index from the non-inferior solution set to obtain a target budget allocation combination;
alternatively, the first and second electrodes may be,
and selecting the optimal solution of each effect index from the non-inferior solution set to obtain a target budget allocation combination.
In another possible implementation manner of the present application, the solutions in the non-inferior solution set are the solutions remaining after the solution with the worst effect index is removed.
In another possible implementation manner of the present application, the obtaining at least one budget allocation combination corresponding to each delivery channel according to the budget total includes:
and obtaining at least one budget allocation combination corresponding to each delivery channel according to the total budget and the constraint conditions of each delivery channel.
In another possible implementation manner of the present application, the historical delivery data includes delivery channel categories and budget allocations; the effect data comprises advertisement display amount, click rate and conversion amount.
The present application further provides a computer program product adapted to perform a program for initializing the following method steps when executed on a data processing device:
performing fitting prediction according to historical release data and effect data corresponding to each release channel respectively to obtain at least one index prediction model corresponding to each release channel;
obtaining at least one budget allocation combination corresponding to each delivery channel according to the budget total, wherein each budget allocation combination comprises the budget allocation amount corresponding to each delivery channel;
for any budget allocation combination, respectively predicting each effect index predicted value corresponding to the current budget allocation amount of the corresponding delivery channel by using an index prediction model corresponding to each delivery channel, wherein the effect index is an index for representing the advertisement delivery effect;
optimizing and solving effect index predicted values corresponding to all budget allocation combinations by using a genetic algorithm to obtain a non-inferior solution set, wherein each solution in the non-inferior solution set is a budget allocation combination corresponding to each delivery channel;
and determining a target budget allocation combination from the non-inferior solution set.
In one possible implementation manner of the present application, the optimizing and solving the effect index predicted values corresponding to all budget allocation combinations by using a genetic algorithm to obtain a non-inferior solution set includes:
setting an optimization target according to the service requirement, and setting a variation probability;
selecting at least two gene sequences according to the optimization target, and performing gene evolution operation on the selected gene sequences according to the variation probability to generate a next generation gene sequence;
carrying out the gene evolution operation on each generation of gene sequence until the iteration times reach the specified iteration times to obtain the non-inferior solution set;
and the corresponding gene segments of all the delivery channels form a gene sequence.
In another possible implementation manner of the present application, the fitting prediction is performed on historical delivery data and effect data corresponding to each delivery channel, so as to obtain at least one index prediction model corresponding to each delivery channel, including:
the following steps are performed for one delivery channel:
acquiring a consumption amount and effect data corresponding to any effect index from historical release data corresponding to the release channel;
and according to the consumption amount and the effect data corresponding to the effect index, performing fitting prediction to obtain an index prediction model corresponding to the effect index.
In another possible implementation manner of the present application, determining a target budget allocation combination from the non-inferior solution set includes:
selecting a solution meeting the overall constraint condition of each effect index from the non-inferior solution set to obtain a target budget allocation combination;
alternatively, the first and second electrodes may be,
and selecting the optimal solution of each effect index from the non-inferior solution set to obtain a target budget allocation combination.
In another possible implementation manner of the present application, the solutions in the non-inferior solution set are the solutions remaining after the solution with the worst effect index is removed.
In another possible implementation manner of the present application, the obtaining at least one budget allocation combination corresponding to each delivery channel according to the budget total includes:
and obtaining at least one budget allocation combination corresponding to each delivery channel according to the budget total and the constraint conditions of each delivery channel.
In another possible implementation manner of the present application, the historical delivery data includes delivery channel categories and budget allocations; the effect data comprises advertisement display amount, click rate and conversion amount.
As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
In a typical configuration, a computing device includes one or more processors (CPUs), input/output interfaces, network interfaces, and memory.
The memory may include forms of volatile memory in a computer readable medium, Random Access Memory (RAM) and/or non-volatile memory, such as Read Only Memory (ROM) or flash memory (flash RAM). The memory is an example of a computer-readable medium.
Computer-readable media, including both non-transitory and non-transitory, removable and non-removable media, may implement information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of the storage medium of the computer include, but are not limited to, phase change memory (PRAM), Static Random Access Memory (SRAM), Dynamic Random Access Memory (DRAM)
(DRAM), other types of Random Access Memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), Digital Versatile Disks (DVD) or other optical storage, magnetic cassettes, magnetic tape storage or other magnetic storage devices, or any other non-transmission medium, which can be used to store information that can be accessed by a computing device. As defined herein, a computer readable medium does not include a transitory computer readable medium such as a modulated data signal and a carrier wave.
It should also be noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in the process, method, article, or apparatus that comprises the element.
As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The above are merely examples of the present application and are not intended to limit the present application. Various modifications and changes may occur to those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of the claims of the present application.
Claims (9)
1. An advertisement placement budget allocation method, comprising:
performing fitting prediction according to historical release data and effect data corresponding to each release channel respectively to obtain a plurality of index prediction models corresponding to each release channel;
obtaining a plurality of budget allocation combinations corresponding to each delivery channel according to the budget total, wherein each budget allocation combination comprises the budget allocation amount corresponding to each delivery channel;
for any budget allocation combination, respectively predicting each effect index predicted value corresponding to the current budget allocation amount of the corresponding delivery channel by using an index prediction model corresponding to each delivery channel, wherein the effect index is an index for representing the advertisement delivery effect;
optimizing and solving effect index predicted values corresponding to all budget allocation combinations by using a genetic algorithm to obtain a non-inferior solution set, wherein each solution in the non-inferior solution set is a budget allocation combination corresponding to each delivery channel;
determining a target budget allocation combination from the non-inferior solution set;
respectively carrying out fitting prediction on historical delivery data and effect data corresponding to each delivery channel to obtain a plurality of index prediction models corresponding to each delivery channel, and the method comprises the following steps:
the following steps are performed for one delivery channel:
acquiring a consumption amount and effect data corresponding to any effect index from historical release data corresponding to the release channel;
and according to the consumption amount and the effect data corresponding to the effect index, performing fitting prediction to obtain an index prediction model corresponding to the consumption amount and the effect index.
2. The method of claim 1, wherein the optimizing and solving the effect index predicted values corresponding to all budget allocation combinations by using a genetic algorithm to obtain a non-inferior solution set comprises:
setting an optimization target according to the service requirement, and setting a variation probability;
selecting at least two gene sequences according to the optimization target, and performing gene evolution operation on the selected gene sequences according to the variation probability to generate a next generation gene sequence;
carrying out the gene evolution operation on each generation of gene sequence until the iteration times reach the specified iteration times to obtain the non-inferior solution set;
and the corresponding effect index predicted value and any budget allocation amount of any delivery channel form a gene segment, and the gene segments corresponding to all the delivery channels form a gene sequence.
3. The method of claim 1, wherein determining a target budget allocation combination from the set of non-inferior solutions comprises:
selecting a solution meeting the overall constraint condition of each effect index from the non-inferior solution set to obtain a target budget allocation combination;
alternatively, the first and second electrodes may be,
and selecting the optimal solution of each effect index from the non-inferior solution set to obtain a target budget allocation combination.
4. The method of claim 1, wherein the solutions in the set of non-inferior solutions are the remaining solutions after removing the solution with the worst performance indicator.
5. The method according to claim 1, wherein the obtaining at least one budget allocation combination corresponding to each delivery channel according to the budget total includes:
and obtaining at least one budget allocation combination corresponding to each delivery channel according to the budget total and the constraint conditions of each delivery channel.
6. The method of claim 1, wherein the historical delivery data includes delivery channel categories and budget allocations; the effect data comprises advertisement display amount, click rate and conversion amount.
7. An advertisement placement budget allocation apparatus, comprising:
the fitting prediction module is used for performing fitting prediction according to historical delivery data and effect data corresponding to each delivery channel respectively to obtain a plurality of index prediction models corresponding to each delivery channel;
the budget allocation module is used for acquiring various budget allocation combinations corresponding to each delivery channel according to the budget total amount, and each budget allocation combination comprises the budget allocation amount corresponding to each delivery channel;
the forecasting module is used for forecasting each effect index forecasting value corresponding to the current budget allocation amount of the corresponding delivery channel by using the index forecasting model corresponding to each delivery channel aiming at any budget allocation combination, and the effect index is an index for representing the advertisement delivery effect;
the optimization module is used for carrying out optimization solution on the effect index predicted values corresponding to all the budget allocation combinations by utilizing a genetic algorithm to obtain a non-inferior solution set, wherein each solution in the non-inferior solution set is a budget allocation combination corresponding to each delivery channel;
a determining module, configured to determine a target budget allocation combination from the non-inferior solution set;
the fitting prediction module respectively performs fitting prediction on historical delivery data and effect data corresponding to each delivery channel to obtain a plurality of index prediction models corresponding to each delivery channel, and the fitting prediction module comprises:
the following steps are performed for one delivery channel:
acquiring a consumption amount and effect data corresponding to any effect index from historical release data corresponding to the release channel;
and according to the consumption amount and the effect data corresponding to the effect index, performing fitting prediction to obtain an index prediction model corresponding to the consumption amount and the effect index.
8. A storage medium on which a program is stored, the program implementing the advertisement placement budget allocation method according to any one of claims 1 to 6 when executed by a processor.
9. A processor for executing a program, wherein the program executes to perform the method for allocating an advertisement placement budget according to any one of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810502118.2A CN110533437B (en) | 2018-05-23 | 2018-05-23 | Advertisement delivery budget allocation method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810502118.2A CN110533437B (en) | 2018-05-23 | 2018-05-23 | Advertisement delivery budget allocation method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110533437A CN110533437A (en) | 2019-12-03 |
| CN110533437B true CN110533437B (en) | 2022-09-20 |
Family
ID=68656523
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810502118.2A Active CN110533437B (en) | 2018-05-23 | 2018-05-23 | Advertisement delivery budget allocation method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110533437B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111652649B (en) * | 2020-06-03 | 2024-01-30 | 广州市丰申网络科技有限公司 | Advertisement targeted delivery method, system, device and storage medium |
| CN113379475B (en) * | 2021-08-04 | 2022-10-11 | 北京达佳互联信息技术有限公司 | Object delivery method, device and storage medium |
| CN114581128A (en) * | 2022-02-28 | 2022-06-03 | 北京沃东天骏信息技术有限公司 | Advertisement putting method and device, electronic equipment and computer readable medium |
| CN114862472A (en) * | 2022-05-19 | 2022-08-05 | 上海钧正网络科技有限公司 | File generation and delivery method, file generation and delivery device, electronic equipment and medium |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102663616A (en) * | 2012-03-19 | 2012-09-12 | 北京国双科技有限公司 | Method and system for measuring web advertising effectiveness based on multiple-contact attribution model |
| CN103136696A (en) * | 2013-03-26 | 2013-06-05 | 明日互动（北京）广告传媒有限公司 | Management method of media placement and system thereof |
| CN104851023A (en) * | 2015-05-07 | 2015-08-19 | 容一飞 | Real-time bidding online feedback control method and system |
| CN106204087A (en) * | 2015-05-06 | 2016-12-07 | 北京派择网络科技有限公司 | For the method and apparatus selecting advertising media |
| CN106875202A (en) * | 2015-12-11 | 2017-06-20 | 北京国双科技有限公司 | Assess the method and device of advertisement delivery effect |
| CN107330725A (en) * | 2017-06-29 | 2017-11-07 | 北京酷云互动科技有限公司 | Advertisement value appraisal procedure, budget allocation method, input appraisal procedure and system |
| CN107481062A (en) * | 2017-08-21 | 2017-12-15 | 小草数语（北京）科技有限公司 | The distribution method and device of advertisement putting budget |
| CN108256893A (en) * | 2016-12-29 | 2018-07-06 | 北京国双科技有限公司 | The analysis method and device of advertisement delivery effect |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20070143186A1 (en) * | 2005-12-19 | 2007-06-21 | Jeff Apple | Systems, apparatuses, methods, and computer program products for optimizing allocation of an advertising budget that maximizes sales and/or profits and enabling advertisers to buy media online |
-
2018
- 2018-05-23 CN CN201810502118.2A patent/CN110533437B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102663616A (en) * | 2012-03-19 | 2012-09-12 | 北京国双科技有限公司 | Method and system for measuring web advertising effectiveness based on multiple-contact attribution model |
| CN103136696A (en) * | 2013-03-26 | 2013-06-05 | 明日互动（北京）广告传媒有限公司 | Management method of media placement and system thereof |
| CN106204087A (en) * | 2015-05-06 | 2016-12-07 | 北京派择网络科技有限公司 | For the method and apparatus selecting advertising media |
| CN104851023A (en) * | 2015-05-07 | 2015-08-19 | 容一飞 | Real-time bidding online feedback control method and system |
| CN106875202A (en) * | 2015-12-11 | 2017-06-20 | 北京国双科技有限公司 | Assess the method and device of advertisement delivery effect |
| CN108256893A (en) * | 2016-12-29 | 2018-07-06 | 北京国双科技有限公司 | The analysis method and device of advertisement delivery effect |
| CN107330725A (en) * | 2017-06-29 | 2017-11-07 | 北京酷云互动科技有限公司 | Advertisement value appraisal procedure, budget allocation method, input appraisal procedure and system |
| CN107481062A (en) * | 2017-08-21 | 2017-12-15 | 小草数语（北京）科技有限公司 | The distribution method and device of advertisement putting budget |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110533437A (en) | 2019-12-03 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110533437B (en) | Advertisement delivery budget allocation method and device | |
| US11200592B2 (en) | Simulation-based evaluation of a marketing channel attribution model | |
| Heilig et al. | A cloud brokerage approach for solving the resource management problem in multi-cloud environments | |
| US20180225588A1 (en) | Accelerated machine learning optimization strategy to determine high performance cluster with minimum resources | |
| US20180225587A1 (en) | Fast multi-step optimization technique to determine high performance cluster | |
| US20130332249A1 (en) | Optimal supplementary award allocation | |
| EP3268918A1 (en) | Auto-expanding campaign optimization | |
| CN103778474A (en) | Resource load capacity prediction method, analysis prediction system and service operation monitoring system | |
| US20180176148A1 (en) | Method of dynamic resource allocation for public clouds | |
| Anders et al. | Cooperative resource allocation in open systems of systems | |
| CN109583921A (en) | Advertising budget acquisition methods, device, storage medium and processor | |
| CN109389424B (en) | Flow distribution method and device, electronic equipment and storage medium | |
| Han et al. | Budget allocation as a multi-agent system of contextual & continuous bandits | |
| US10313457B2 (en) | Collaborative filtering in directed graph | |
| US20230289847A1 (en) | Machine learning with data synthesization | |
| CN113342418A (en) | Distributed machine learning task unloading method based on block chain | |
| CN111047435A (en) | Credit data processing method, credit allocation method, credit data processing device, credit allocation device and electronic equipment | |
| CN116739665A (en) | Information delivery method and device, electronic equipment and storage medium | |
| CN115345635A (en) | Processing method and device for recommended content, computer equipment and storage medium | |
| CN109146536A (en) | A kind of information is bidded the method and apparatus of operation | |
| US11281983B2 (en) | Multi-agent system for efficient decentralized information aggregation by modeling other agents' behavior | |
| Zhang et al. | A Dynamic Pricing Model for Virtual Machines in Cloud Environments | |
| Afshar et al. | Dynamic Ad Network Ordering Method Using Reinforcement Learning | |
| CN113393330B (en) | Financial wind control management system based on block chain | |
| CN111291894B (en) | Resource scheduling method, device, equipment and medium in super-parameter optimization process |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |