CN114693343A - Advertisement budget allocation method, device, equipment and storage medium - Google Patents
Advertisement budget allocation method, device, equipment and storage medium Download PDFInfo
- Publication number
- CN114693343A CN114693343A CN202210249490.3A CN202210249490A CN114693343A CN 114693343 A CN114693343 A CN 114693343A CN 202210249490 A CN202210249490 A CN 202210249490A CN 114693343 A CN114693343 A CN 114693343A
- Authority
- CN
- China
- Prior art keywords
- budget
- value
- contact
- contacts
- constraint condition
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 41
- 238000003860 storage Methods 0.000 title claims abstract description 8
- 238000004590 computer program Methods 0.000 claims description 10
- 230000009467 reduction Effects 0.000 claims description 6
- 238000005457 optimization Methods 0.000 description 20
- 230000006870 function Effects 0.000 description 13
- 238000004364 calculation method Methods 0.000 description 9
- 238000004422 calculation algorithm Methods 0.000 description 7
- 230000008901 benefit Effects 0.000 description 6
- 239000002245 particle Substances 0.000 description 6
- 238000010586 diagram Methods 0.000 description 4
- 238000004458 analytical method Methods 0.000 description 3
- 230000008859 change Effects 0.000 description 3
- 230000003247 decreasing effect Effects 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 238000009987 spinning Methods 0.000 description 3
- 238000006243 chemical reaction Methods 0.000 description 2
- 230000001419 dependent effect Effects 0.000 description 2
- 238000002156 mixing Methods 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 230000001133 acceleration Effects 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 230000003542 behavioural effect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000002068 genetic effect Effects 0.000 description 1
- 230000013016 learning Effects 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000001737 promoting effect Effects 0.000 description 1
- 230000009326 social learning Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012549 training Methods 0.000 description 1
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
- G06Q30/0249—Advertisements based upon budgets or funds
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2111/00—Details relating to CAD techniques
- G06F2111/04—Constraint-based CAD
Abstract
The application discloses an advertisement budget allocation method, an advertisement budget allocation device, advertisement budget allocation equipment and a storage medium. The method comprises the following steps: if the solution result of the preset budget allocation model does not meet the constraint condition of the budget allocation model, determining the contact which does not meet the constraint condition and/or a contact group consisting of at least two contacts to obtain a target object; adjusting the budget of the target object, wherein the target object after the budget is adjusted meets the constraint condition; updating the solution result according to the adjusted budget; and outputting the updated solving result. The scheme provided by the application aims to solve the technical problem that the solution efficiency of the budget allocation model in the prior art is low.
Description
Technical Field
The present application relates to the field of information, and in particular, to a method, an apparatus, a device, and a storage medium for allocating an advertisement budget.
Background
Advertisement putting is an online marketing mode, is widely popularized by advertisers due to the advantages of simple operation, controllable budget, high advertisement benefit and the like, is one of main marketing means of the present generation, and is also one of hot spots of recent researches of scholars at home and abroad. The advertiser can put advertisements through online advertisements, and sets a corresponding budget for each contact media, so that the advertisement profit maximization becomes the most concerned research content of the advertiser under the preset advertisement budget. Advertisement budget allocation is a primary problem that must be solved in keyword bidding promotional activities. A reasonable budget allocation scheme can optimize the campaign "top-down" so as to preempt in intense competition. The nature of advertisement budget allocation is an operation optimization problem for solving linear or non-linear programming, and optimal budget allocation is converted into an optimal solution of a programming problem.
In the related technology, by obtaining advertisement delivery return information of each advertiser, such as exposure, advertisement information, advertisement positions and other information, unknown parameters of the access rate of each contact are obtained through a least square fitting method, an optimization model is constructed through analyzing business rules, a target function and corresponding constraint conditions are set, solving operation is executed, and the exposure rate corresponding to each contact is obtained.
In practical application, the budget allocation model is a nonlinear mathematical optimization model, and the nonlinear optimization model is an optimization model which is difficult to solve, so that the problem of low solution efficiency of the budget allocation model is caused.
Disclosure of Invention
The application mainly aims to provide an advertisement budget allocation method, an advertisement budget allocation device, advertisement budget allocation equipment and a storage medium, and aims to solve the technical problem that in the prior art, a budget allocation model is low in solving efficiency.
In order to achieve the above object, the present application proposes the following technical solutions, including:
an advertising budget allocation method, comprising:
if the solution result of the preset budget allocation model does not meet the constraint condition of the budget allocation model, determining the contact which does not meet the constraint condition and/or a contact group consisting of at least two contacts to obtain a target object;
adjusting the budget of the target object, wherein the target object after the budget is adjusted meets the constraint condition;
updating the solution result according to the adjusted budget;
and outputting the updated solving result.
An advertising budget allocation apparatus, comprising:
the determining module is used for determining the contact which does not meet the constraint condition and/or a contact group consisting of at least two contacts if the solving result of the preset budget allocation model does not meet the constraint condition of the budget allocation model, so as to obtain a target object;
the adjusting module is used for adjusting the budget of the target object, wherein the target object after the budget is adjusted meets the constraint condition;
the updating module is used for updating the solving result according to the adjusted budget;
and the output module is used for outputting the updated solving result.
An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the method described above when executing the computer program.
A computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the method as described above.
In the technical scheme of the application, if the solution result of a preset budget allocation model does not meet the constraint condition of the budget allocation model, determining a contact which does not meet the constraint condition and/or a contact group consisting of at least two contacts to obtain a target object, adjusting the budget of the target object, wherein the budget-adjusted target object meets the constraint condition, updating the solution result according to the adjusted budget, and outputting the updated solution result; by adjusting the budget of the target object, the solving result can meet the constraint condition of the budget allocation model, the nonlinear mathematical optimization model can obtain the solving result meeting the constraint condition as soon as possible, and the calculation efficiency is improved.
Drawings
In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present application, and for those skilled in the art, other drawings can be obtained according to the structures shown in the drawings without creative efforts.
FIG. 1 is a flowchart of an advertisement budget allocation method according to an embodiment of the present application;
FIG. 2 is a schematic diagram illustrating adjustment of a total budget constraint provided by an embodiment of the present application;
FIG. 3 is a flowchart of a method for adjusting a budget based on constraints of a contact set according to an embodiment of the present application;
fig. 4 is a schematic diagram of an advertisement budget allocation apparatus according to an embodiment of the present application.
Detailed Description
In order to make the objects, technical solutions and advantages of the present application more apparent, the present application will be described and illustrated below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of and not restrictive on the broad application. All other embodiments obtained by a person of ordinary skill in the art based on the embodiments provided in the present application without any inventive step are within the scope of protection of the present application.
It is obvious that the drawings in the following description are only examples or embodiments of the present application, and that it is also possible for a person skilled in the art to apply the present application to other similar contexts on the basis of these drawings without inventive effort. Moreover, it should be appreciated that in the development of any such actual implementation, as in any engineering or design project, numerous implementation-specific decisions must be made to achieve the developers' specific goals, such as compliance with system-related and business-related constraints, which may vary from one implementation to another.
Reference in the specification to "an embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment can be included in at least one embodiment of the specification. The appearances of the phrase in various places in the specification are not necessarily all referring to the same embodiment, nor are separate or alternative embodiments mutually exclusive of other embodiments. Those of ordinary skill in the art will explicitly and implicitly appreciate that the embodiments described herein may be combined with other embodiments without conflict.
Unless defined otherwise, technical or scientific terms referred to herein shall have the ordinary meaning as understood by those of ordinary skill in the art to which this application belongs. Reference to "a," "an," "the," and similar words throughout this application are not to be construed as limiting in number, and may refer to the singular or the plural. The use of the terms "including," "comprising," "having," and any variations thereof herein, is meant to cover a non-exclusive inclusion; for example, a process, method, system, article, or apparatus that comprises a list of steps or modules (elements) is not limited to the listed steps or elements, but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus. Reference to "connected," "coupled," and the like in this application is not intended to be limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. The term "plurality" as referred to herein means two or more. "and/or" describes an association relationship of associated objects, meaning that three relationships may exist, for example, "A and/or B" may mean: a exists alone, A and B exist simultaneously, and B exists alone. The character "/" generally indicates that the former and latter associated objects are in an "or" relationship. Reference herein to the terms "first," "second," "third," and the like, are merely to distinguish similar objects and do not denote a particular ordering for the objects.
The present application is described in detail with reference to the embodiments shown in the drawings, but it should be understood that these embodiments are not intended to limit the present application, and those skilled in the art should understand that the functional, methodological, or structural equivalents or substitutions made by these embodiments are within the scope of the present application.
Before describing in detail various embodiments of the present application, the core inventive concepts of the present application are summarized and described in detail by the following embodiments.
Fig. 1 is a flowchart of an advertisement budget allocation method according to an embodiment of the present application. As shown in fig. 1, the method includes:
the Touch Point (TP) includes dimensions such as a region, a media, and an advertisement type.
In one exemplary embodiment, the constraints may be constraints for all contacts, or constraints for a single contact, or constraints for groups of contacts.
In an exemplary embodiment, the constraint condition is determined according to at least one of budget, exposure and reach rate, wherein exposure (imcompression, abbreviated as imp) refers to a display behavior of an advertisement, exposure of a certain contact refers to a sum of exposure times of the contact, reach rate (reach), and a proportion of a target user group which can be reached by a channel advertisement delivery.
For example, limiting the budget of all contacts to a certain value range is a constraint condition for all contacts; the exposure of a certain contact is limited in a certain value range, and is a constraint condition aiming at a single contact; limiting the reach of a contact set to be greater than a threshold is a constraint for the contact set.
Wherein a plurality of constraints may be set for all contacts, for example, corresponding constraints may be set according to at least two of budget, exposure and reach respectively; alternatively, there may be one or more constraints for the same contact or group of contacts; for example, for a certain contact group, the exposure of the contact group is limited to a certain value range, and the budget of the contact group is also limited to another value range.
In this step, the object to be adjusted can be determined by determining the target object, so that the solution result after adjustment can meet the constraint condition of the budget allocation model.
because the constraint conditions of the target objects are all determined by the respective budget sizes, if the budget of the target object exceeds the reasonable budget corresponding to the constraint conditions, the range exceeding the limit of the constraint conditions can appear; on the contrary, if the budget of the target object does not reach the reasonable budget corresponding to the constraint condition, the range limited by the constraint condition is not reached enough.
Based on the analysis, the budget of the target object is adjusted to be within a reasonable range corresponding to the constraint condition, so that the respective constraint condition is met.
103, updating the solution result according to the adjusted budget;
the value of the solution result changes due to the adjustment of the budget of the target object, and therefore, the solution result needs to be updated.
Taking the solution result as the exposure amount for each contact point as an example, the exposure amount for a single contact point or all or part of the contact points in the contact point group may also change because the budget of the contact point and/or the contact point group as the target object is adjusted.
Further, after the budget is adjusted, all the contact points of the target object meet respective constraint conditions, so the updated solution results determined based on the adjusted budget also meet the constraint conditions of the budget allocation model.
104, outputting the updated solving result;
in the method provided by the embodiment of the application, if the solution result of the preset budget allocation model does not meet the constraint condition of the budget allocation model, determining the contact which does not meet the constraint condition and/or the contact group consisting of at least two contacts to obtain the target object, adjusting the budget of the target object, wherein the budget-adjusted target object meets the constraint condition, updating the solution result according to the adjusted budget, and outputting the updated solution result; by adjusting the budget of the target object, the solution result can meet the constraint condition of the budget allocation model, so that the nonlinear mathematical optimization model can obtain the solution result meeting the constraint condition as soon as possible, and the calculation efficiency is improved.
The method provided by the embodiments of the present application is explained as follows:
the budget allocation aims at advertisement delivery based on reach prediction and allocation, and is applied to a third-party advertisement company for helping an advertiser to decide, wherein the advertiser can be helped to decide how much budget to deliver in each channel in advance so as to obtain the maximum reach, and the advertiser can be helped intelligently to decide how to deliver the advertisements through the advertisement budget allocation so as to make the best use of the best of the materials, so that the maximum reach is obtained. Meanwhile, the intelligent decision-making can be carried out on the putting of different platforms in advance from the perspective of an advertiser before the advertisement is put, and the benefits generated by the intelligent decision-making can be estimated, so that the advertisement putting of the advertiser is guided and is not blind, and the attention is paid to the number of the advertisers.
First, a budget allocation model for implementing an advertisement budget allocation is obtained by:
a1, obtaining advertisement putting return information of each advertiser, such as exposure, advertisement information, advertisement position and other information;
and A2, processing the advertisement putting return information through a preset business analysis rule, and constructing an optimization model. And establishing a budget allocation objective function (the sum of the reaching rates of all the contacts is maximum) and corresponding constraint conditions (constraint limits such as medium budget of each contact, total budget and the like).
And A3, solving by adopting a solver to obtain an optimal solution meeting the constraint so as to solve the problem of the solving efficiency of the optimization model.
The solver can be a solver of a Particle swarm optimization algorithm (PSO).
In step a2, the unknown parameters of the contact medium arrival rate are obtained by the least square fitting method. It is mathematically transformed by analyzing the business rules and an optimization model is constructed. Constructing a corresponding budget allocation model based on the optimization model, wherein the objective function and the constraint condition of the budget allocation model,
the optimization model is specifically as follows:
wherein a and b are real-time and x is a variable.
By obtaining advertisement putting return information of each advertiser, such as exposure, market information, advertisement positions and other information, wherein the most important field is the exposure, when the exposure is larger, the contact conversion rate is also larger, but the marginal benefit problem exists, and when the exposure is larger to a certain degree, the contact conversion rate is approximately close to a certain asymptote. The advertiser refers to a group who wants to put an advertisement.
The objective function may be to maximize reach:
the constraints include at least one of:
the total expenditure of all contacts is not allowed to exceed the total budget:
for some contact groups, add exposure upper and lower bounds constraints:
for some groups of contacts, add upper and lower bounds constraints on cost:
xij: exposure allocated for each advertising spot TP, as a decision variable
Pij: price per TP Unit exposure
groupk: the kth group contains decision variables of constraints of the same type
g (x): exposure-to-touch rate function, using Google function
Its independent variable is the exposure and the dependent variable is the reach produced by the exposure.
f (x): de-duplication function, taking the exponential function f as 1-eaxThe independent variable is reach before deduplication, and the dependent variable is reach after deduplication. The purpose of deduplication is to eliminate duplicate groups, that is, duplicate users in different channels are eliminated, for example, user a watches the same advertisement in both a mobile phone and a computer, and then for reach calculation, the two exposures reach the same individual, and are therefore duplicate.
And (3) budget: the total allocated budget, i.e., the total budget that the advertiser wants to allocate this time.
imp: exposure, i.e. the sum of the exposures of some point does not exceed a certain value.
spinning: the sum of the costs, i.e. of some points, does not exceed a certain value.
The method is characterized in that modeling is carried out on the advertisement budget allocation problem, the advertisement budget allocation problem is summarized into a mathematical optimization problem, the maximum reach is taken as a target, the cost, the exposure and the budget are taken as constraints, the advertisement budget allocation problem is solved, and the calculation efficiency and the calculation accuracy of the optimal solution are improved.
In addition, in the embodiment of the present application, the revenue information of each contact may be the generated reach rate Δ reach due to the budget Δ budget required for increasing the exposure amount of a fixed size for each contact. That is, the profit information per unit price is obtained by dividing the increased reach rate Δ reach by the required increased budget Δ budget, and the profit information can be represented by value.
Specifically, acquiring the reaching rate and the cost required to pay corresponding to the exposure amount of each contact point increased by the preset size; and calculating the ratio of the reaching rate to the cost of the same contact point to obtain the income information of each contact point.
The calculation expression of the reach rate reach is as follows:
when the budget allocation model is used for performing budget allocation, the reaching rate generated by increasing the unit budget of each contact is calculated, and the contact allocation budget with a high value is preferentially selected until the budget is completely allocated.
Wherein, the calculation expression of value is as follows:
the larger the value is, the higher the investment value is represented; then the priority should be higher when allocating the budget in favor of a larger reach.
In the present application, the object constrained by the constraint condition can be described as two cases, one is a whole contact, and the other is a contact or a contact group.
First, the adjustment method corresponding to the constraint conditions of all the contacts will be described:
in one exemplary implementation, when the constraint is a constraint of all touch points, the target object is determined by:
acquiring income information of each contact in all contacts under unit price; obtaining the comparison result between the total value corresponding to all the contacts and the first upper limit value and the first lower limit value in the constraint conditions of all the contacts;
if the total number value corresponding to all the contacts is smaller than the first lower limit value, selecting a target object from all the contacts according to the sequence from high to low of the income information;
and if the total number value corresponding to all the contacts is larger than the first upper limit value, selecting the target objects from all the contacts according to the sequence from low to high of the profit information.
Correspondingly, when the constraint condition is a constraint condition of all the touch points, the method for adjusting the budget of the target object comprises the following steps:
if the total number of all the contacts is larger than a first upper limit value in the constraint condition of all the contacts, determining a first difference value between the first upper limit value and the total number, and removing a budget with the size of the first difference value from the expenditure of the target object;
and if the total number of all the contacts is smaller than the first lower limit value in the constraint condition of all the contacts, determining a second difference value between the total number and the first lower limit value, and increasing the budget with the second difference value in the expenditure of the target object.
The following is a specific example of the constraint conditions of all the contacts:
the constraints for all contacts are as follows:
the budget sum of all the contacts is smaller than the preset total budget upper limit value and larger than the preset total budget lower limit value.
Since value is the amount of change in the objective function due to an increase in the unit budget, it corresponds to the slope of the tangent on the curve of the objective function. Wherein, the higher the value is, the more income brought by budget investment is represented, and the higher the investment priority is; conversely, the lower the investment priority. Therefore, when the budget of a certain contact is reduced, the contact with a low value is selected as much as possible; when the budget of a certain touch point is increased, the touch point with the value as high as possible is selected.
Based on the analysis, the constraint conditions are processed in the following ways, including:
if the budget sum of all the contacts is between the total budget upper limit value and the total budget lower limit value, the constraint condition is met, and adjustment is not needed;
if the sum of the budgets of all the contacts is greater than the total budget upper limit value, the sum of the budgets of all the contacts is beyond the range of the total budget, one or at least two contacts are selected to reduce the budgets according to the descending order of the value, and the reduced budgets are greater than or equal to the difference between the sum of the budgets and the total budget upper limit value;
and if the sum of the budgets of all the contacts is smaller than the lower limit value of the total budget and indicates that the sum of the budgets of all the contacts does not reach the range of the total budget, selecting one or at least two contacts to increase the budget according to the descending order of the value until the distribution is finished.
The selected contact points are used as target objects, and the number of the target objects is determined according to the total budget amount required to be adjusted.
Further, if the target objects have the budget constraints corresponding to the target objects, when the budget of the target object is adjusted, it is further required to ensure that the adjusted budgets of the target objects still conform to the budget constraints corresponding to the target objects.
The following describes the adjustment method corresponding to the constraint condition of a single contact or the constraint condition of a contact group:
in an exemplary embodiment, when the constraint is a constraint of a single contact or a group of contacts, adjusting the budget of the target object includes:
judging whether the target object comprises a first object and a second object, wherein if the first object meets a constraint condition, the budget with the size of a first value needs to be reduced; if the second object meets the constraint condition, the budget with the size of the second value needs to be increased;
if the target object only has the first object, allocating the budget with the size of the first value to the contact and/or the contact group which meets the constraint condition, wherein the newly added contact and/or the contact group which meets the budget with the size of the first value still meet the respective constraint condition;
if the target object only has a second object, reducing the budget with the second value from the budgets of the contacts and/or the contact groups meeting the constraint condition, and allocating the budget with the second value to the second object, wherein the budgets with the second value and the reduced contacts and/or the contact groups still meet the respective constraint condition;
and if the target object comprises a first object and a second object, performing budget allocation operation according to the comparison result of the first value and the second value.
In the above steps, the target objects are divided into two types, one type is that the target objects need to satisfy the corresponding constraint conditions by reducing the budget, and the other type is that the target objects need to satisfy the corresponding constraint conditions by increasing the budget.
From the above, when the target object has only one of the two types, that is, when there is only the first object or the second object in the target object, it is necessary to assist with the contacts or the contact groups that have satisfied the constraint conditions to complete the budget reallocation, and at the same time, it is necessary to ensure that the contacts and/or the contact groups that have increased the budget or decreased the budget still satisfy the respective constraint conditions.
When the target object has the two types, that is, the target object has the first object and the second object, the budget reallocation is required according to the adjusted budget sizes of the first object and the second object.
Specifically, the performing budget allocation operation according to the comparison result of the first value and the second value includes:
allocating a budget of a first value to the second object if the first value is equal to the second value;
if the first value is larger than the second value and the difference value is a third difference value, allocating the budget with the second value from the budget of the first object to the second object, and allocating the budget with the third difference value to the contact and/or the contact group which meets the preset condition, wherein the contact and/or the contact group which is newly added with the budget with the third difference value still meets the respective constraint condition;
if the first value is less than the second value and the difference is a fourth difference, reducing the budget for the fourth difference from the expenditure of the contact and/or the contact group for which the constraint has been met, and allocating the budget for the fourth difference and the budget for the first value to the second object, wherein the contact and/or the contact group for which the budget for the fourth difference has been reduced still meet the respective constraint.
In this step, if the first value is equal to the second value, the excess budget of the first object can be directly allocated to the second object, so that the first object and the second object both conform to the respective constraints. On the contrary, if the first value is not equal to the second value, the budget may be reallocated by using the contact or the contact group that has satisfied the constraint, and it may be ensured that the contact and/or the contact group that has increased the budget or decreased the budget still satisfies the respective constraint.
Further, when the first object is a contact group, the budget having the size of the first value is obtained by:
acquiring income information of each contact in the contact group;
selecting one or at least two contacts from the contact group according to the sequence of the profit information of each contact in the contact group from low to high, and executing budget reduction operation with the total number of the budget reduction operation being a first value;
and/or the presence of a gas in the gas,
when the second object is a contact group, the budget having the second value is obtained by:
acquiring income information of each contact in the contact group;
and selecting one or at least two contacts from the contact group according to the sequence from high to low of the profit information of each contact in the contact group, and executing the budget increasing operation with the total number of the budget increasing operation being a second value.
The value of the profit information is the change amount of the objective function caused by the increase of the unit budget; wherein, the higher the value is, the more income brought by budget investment is represented, and the higher the investment priority is; conversely, the lower the investment priority. Because the contact set comprises at least two contacts, the contact with the low value is selected as much as possible when the budget of the contact set is reduced; when increasing the budget of the contact set, the contact with the value higher is selected as much as possible.
The following description takes an application scenario as an example:
the solution of the model may be performed using a PSO, wherein the budget allocated to each contact by a decision variable in the PSO-based budget allocation model.
PSO is an evolutionary computing technique. Derived from behavioral studies on predation of groups of birds. The basic idea of the particle swarm optimization algorithm is as follows: the optimal solution is found through cooperation and information sharing among individuals in a group. The PSO algorithm does not require the form of an objective function and a constraint condition, and can solve a complex optimization problem. This has the advantage of being simple and easy to implement and without adjustment of many parameters. The method is widely applied to the application fields of function optimization, neural network training, fuzzy system control and other genetic algorithms.
The decoding mode of the budget allocation model based on the PSO is specifically shown in table 1:
TABLE 1
The parameters mentioned in table 1 are illustrated:
(1) a contact group tppgroup, wherein contacts within the same contact group have the same constraints;
(2) constraint type: belongs to spinning/imp/reach … …, which are respectively numbered 1/2/3 … …; for checking whether the constraints are satisfied;
(3) the constraint satisfies: whether each contact set satisfies a constraint, 1 being a minimum value less than the constraint; 2 is to satisfy the constraint; 3 is a maximum value greater than the constraint;
(4) and (3) Budget: what the budget (budget) of each contact is as a decision variable of the budget allocation model.
Constraints of the budget allocation model:
(1) the total budget cannot exceed a given value;
(2) the constraints of the contact group can set the ending, imp, reach of the contact group to be between the corresponding minimum value min and maximum value max.
The processing flow of the above budget allocation model based on the PSO is as follows:
b1, initial solution generation
The exposure for each touch point is randomly generated within the range of feasible solutions and an objective function is calculated. In the present invention, the initialized exposure is actually the position of the particles.
B2, position and velocity update
Vid＝w*Vid+C1*random(0,1)(Pid-Xid)
+C2*random(0,1)(Pgd-Xid)
Xid＝Xid+Vid
w is an inertia factor, is not negative, and has strong global optimizing capability and weak local optimizing capability when being larger; when the time is small, the global optimizing capability is weak, and the local optimizing capability is strong. By adjusting, global and local optimality can be adjusted. C1 and C2 are acceleration constants, the former being individual learning factors and the latter being social learning factors. Literature studies, with C1 and C2 set to 2 generally, the results were better. But not necessarily equal to 2, and typically between 0 and 4.
B3, constraint adjustment
After the position and velocity updates are made, the constraints may not be satisfied and need to be adjusted by the budget.
The following is a description of different constraints:
1) constraint of total budget:
if the total budget is met, no adjustment is needed;
if the total budget is exceeded, finding out the value which is the minimum, and reducing the investment budget;
if not, allocating the rest budget to the value with the maximum value until the allocation is finished;
fig. 2 is a schematic diagram of adjusting a total budget constraint provided in an embodiment of the present application. As shown in FIG. 2, the budget is increased preferentially for contacts with a value high and decreased preferentially from contacts with a value low.
(2) Spinning, imp, reach constraints per tpgorup
Constraint adjustment is performed according to the "budget constraint adjustment scheme" recorded in table 2, which is as follows:
TABLE 2
In conjunction with table 2, the constraint adjustment procedure is as follows:
calculating the blending and imp constraints of each tpgorup, if both the blending and imp constraints are satisfied, the scheme is feasible without adjustment; if not, the next step is carried out;
adjusting according to all the conditions of table 2 until the constraints are satisfied; all cases where the constraints are not satisfied are shown in table 2.
And setting a stopping rule of constraint adjustment, for example, after the constraint adjustment is carried out for countless times, if all the constraint adjustment is 2, stopping the constraint adjustment.
FIG. 3 is a flowchart of a method for adjusting a budget based on constraints of a contact set according to an embodiment of the present disclosure. As shown in fig. 3, the method includes:
step C1, identifying the constraint satisfaction condition of each contact group, wherein: the condition that the min is smaller than the minimum value min of the constraint condition is marked as 1, the condition that the constraint condition is met is marked as 2, and the condition that the min is larger than the maximum value max of the constraint condition is marked as 3;
step C2, judging whether a contact group with the maximum value larger than the constraint condition exists;
if so, go to step C3; otherwise, go to step C4;
step C3, subtracting partial budget from the budget of the contact with the minimum value from the contact with the minimum value in the contact group larger than the maximum value of the constraint condition according to a preset proportion, and storing the budget in the residual budgetleftIf the minimum value of the constraint condition of the contact group is ensured to be met, continuing to the step C2;
step C4, judging whether a contact group smaller than the minimum value of the constraint condition exists;
if so, performing steps C5-C7;
if not, performing steps C8-C9;
step C5, judging whether the residual budget is 0;
if so, go to step C6; otherwise, executing step C7;
step C6, subtracting partial budget from the budget of the contact with the minimum value from the contact with the minimum value in the contact group meeting the constraint condition according to a preset proportion, and storing the budget in the residual budgetleftIf the minimum value of the constraint condition of the contact group is ensured to be met, continuing to the step C5;
step C7, increasing the budget of the contact with the maximum value according to a preset proportion from the contact with the maximum value in the contact group smaller than the minimum value of the constraint condition, ensuring that the minimum value of the constraint condition of the contact group is met, updating the residual budget, and continuing to the step C5;
step C8, judging whether the residual budget is 0;
if so, go to step C9; otherwise, the flow ends.
Step C6, subtracting partial budget from the budget of the contact with the maximum value in the contact group meeting the constraint condition according to a preset proportion, and storing the budget to the residual budget bucketl tThe minimum and maximum values of the contact set constraints are guaranteed to be met and the process continues to step C8.
From the above, aiming at a nonlinear optimization model which is difficult to solve, the embodiment of the application provides a particle swarm algorithm which can solve a nonlinear optimization problem, and particularly, aiming at the problem that the constraint of the particle swarm algorithm cannot be met, the particle swarm algorithm is adjusted, so that the solution can better meet the constraint. The solving speed is high, and the precision is high.
Fig. 4 is a block diagram of an advertisement budget allocation apparatus according to an embodiment of the present application. As shown in fig. 4, the apparatus includes:
the determining module is used for determining the contact which does not meet the constraint condition and/or the contact group consisting of at least two contacts if the solving result of the preset budget allocation model does not meet the constraint condition of the budget allocation model, so as to obtain a target object;
the adjusting module is used for adjusting the budget of the target object, wherein the target object after the budget is adjusted meets the constraint condition;
the updating module is used for updating the solving result according to the adjusted budget;
and the output module is used for outputting the updated solving result.
In one exemplary embodiment, the constraint is determined according to at least one of budget, exposure, and reach.
In an exemplary embodiment, the determining module is specifically configured to determine, when the constraint condition is a constraint condition of all contact points, the target object by:
acquiring income information of each contact in all contacts under unit price; obtaining the comparison result between the total value corresponding to all the contacts and the first upper limit value and the first lower limit value in the constraint conditions of all the contacts;
if the total number value corresponding to all the contacts is smaller than the first lower limit value, selecting a target object from all the contacts according to the sequence from high to low of the income information;
and if the total number value corresponding to all the contacts is larger than the first upper limit value, selecting the target objects from all the contacts according to the sequence from low to high of the profit information.
In an exemplary embodiment, the adjusting module is specifically configured to, when the constraint condition is a constraint condition of all contact points, adjust the budget of the target object by:
if the total number of all the contacts is larger than a first upper limit value in the constraint condition of all the contacts, determining a first difference value between the first upper limit value and the total number, and removing a budget with the size of the first difference value from the expenditure of the target object;
and if the total number of all the contacts is smaller than the first lower limit value in the constraint condition of all the contacts, determining a second difference value between the total number and the first lower limit value, and increasing the budget with the second difference value in the expenditure of the target object.
In an exemplary embodiment, the adjusting module is specifically configured to, when the constraint is a constraint of a single contact or a contact group, adjust the budget of the target object by:
judging whether the target object comprises a first object and a second object, wherein if the first object meets a constraint condition, the budget with the size of a first value needs to be reduced; if the second object meets the constraint condition, the added amount is the budget of a second value;
if the target object only has the first object, allocating the budget with the size of the first value to the contact and/or the contact group which meets the constraint condition, wherein the newly added contact and/or the contact group which meets the budget with the size of the first value still meet the respective constraint condition;
if the target object only has a second object, reducing the budget with the second value from the budgets of the contacts and/or the contact groups meeting the constraint condition, and allocating the budget with the second value to the second object, wherein the budgets with the second value and the reduced contacts and/or the contact groups still meet the respective constraint condition;
and if the target object comprises a first object and a second object, performing budget allocation operation according to the comparison result of the first value and the second value.
In an exemplary embodiment, the performing the budget allocation operation according to the comparison result of the first value and the second value includes:
allocating a budget of a first value to the second object if the first value is equal to the second value;
if the first value is larger than the second value and the difference value is a third difference value, allocating the budget with the second value from the budget of the first object to the second object, and allocating the budget with the third difference value to the contact and/or the contact group which meets the preset condition, wherein the contact and/or the contact group which is newly added with the budget with the third difference value still meets the respective constraint condition;
if the first value is less than the second value and the difference is a fourth difference, reducing the budget for the fourth difference from the expenditure of the contact and/or the contact group for which the constraint has been met, and allocating the budget for the fourth difference and the budget for the first value to the second object, wherein the contact and/or the contact group for which the budget for the fourth difference has been reduced still meet the respective constraint.
In an exemplary embodiment, when the first object is a contact group, the budget having the first value is obtained by:
acquiring income information of each contact in the contact group;
selecting one or at least two contacts from the contact group according to the sequence of the profit information of each contact in the contact group from low to high, and executing budget reduction operation with the total number of the budget reduction operation being a first value;
and/or the presence of a gas in the gas,
when the second object is a contact group, the budget having the second value is obtained by:
acquiring income information of each contact in the contact group;
and selecting one or at least two contact points from the contact point group according to the sequence from high to low of the profit information of each contact point in the contact point group, and executing the budget increase operation with the total number of the budget increase operation as a second value.
According to the device provided by the embodiment of the application, if the solving result of the preset budget allocation model does not meet the constraint condition of the budget allocation model, the contact which does not meet the constraint condition and/or the contact group consisting of at least two contacts are determined to obtain the target object, the budget of the target object is adjusted, wherein the budgeted target object meets the constraint condition, the solving result is updated according to the adjusted budget, and the updated solving result is output; by adjusting the budget of the target object, the solution result can meet the constraint condition of the budget allocation model, so that the nonlinear mathematical optimization model can obtain the solution result meeting the constraint condition as soon as possible, and the calculation efficiency is improved.
An embodiment of the present application provides an electronic device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, and when the processor executes the computer program, the method described above is implemented.
An embodiment of the present application provides a computer-readable storage medium, on which a computer program is stored, where the computer program is configured to implement the method described in any one of the above when executed by a processor.
The above description is only a preferred embodiment of the present application, and not intended to limit the scope of the present application, and all modifications and equivalents of the subject matter of the present application, which is conceived to be equivalent to the above description and the accompanying drawings, or to be directly/indirectly applied to other related arts, are intended to be included within the scope of the present application.
Claims (10)
1. An advertisement budget allocation method, comprising:
if the solution result of the preset budget allocation model does not meet the constraint condition of the budget allocation model, determining the contact which does not meet the constraint condition and/or a contact group consisting of at least two contacts to obtain a target object;
adjusting the budget of the target object, wherein the target object after the budget is adjusted meets the constraint condition;
updating the solution result according to the adjusted budget;
and outputting the updated solving result.
2. The method of claim 1, wherein the constraints are determined based on at least one of budget, exposure, and reach.
3. The method of claim 1, wherein when the constraint is a constraint of all touch points, determining the target object by:
acquiring income information of each contact in all contacts under unit price; obtaining the comparison result between the total value corresponding to all the contacts and the first upper limit value and the first lower limit value in the constraint conditions of all the contacts;
if the total number value corresponding to all the contacts is smaller than the first lower limit value, selecting a target object from all the contacts according to the sequence from high to low of the income information;
and if the total number of all the contact points is greater than the first upper limit value, selecting the target objects from all the contact points according to the sequence from low profit information to high profit information.
4. The method of any one of claims 1 to 3, wherein adjusting the budget of the target object when the constraint is a constraint of all touch points comprises:
if the total number of all the contacts is larger than a first upper limit value in the constraint condition of all the contacts, determining a first difference value between the first upper limit value and the total number, and removing a budget with the size of the first difference value from the expenditure of the target object;
and if the total number of all the contacts is smaller than the first lower limit value in the constraint condition of all the contacts, determining a second difference value between the total number and the first lower limit value, and increasing the budget with the second difference value in the expenditure of the target object.
5. The method of claim 1 or 2, wherein adjusting the budget of the target object when the constraint is a constraint of a single contact or a group of contacts comprises:
judging whether the target object comprises a first object and a second object, wherein if the first object meets a constraint condition, the budget with the size of a first value needs to be reduced; if the second object meets the constraint condition, the added amount is the budget of a second value;
if the target object only has the first object, allocating the budget with the size of the first value to the contact and/or the contact group which meets the constraint condition, wherein the newly added contact and/or the contact group which meets the budget with the size of the first value still meet the respective constraint condition;
if the target object only has a second object, reducing the budget with the second value from the budgets of the contacts and/or the contact groups meeting the constraint condition, and allocating the budget with the second value to the second object, wherein the budgets with the second value and the reduced contacts and/or the contact groups still meet the respective constraint condition;
and if the target object comprises a first object and a second object, performing budget allocation operation according to the comparison result of the first value and the second value.
6. The method of claim 5, wherein performing the budget allocation operation based on the comparison of the first and second values comprises:
allocating a budget of a first value to the second object if the first value is equal to the second value;
if the first value is larger than the second value and the difference value is a third difference value, allocating the budget with the second value from the budget of the first object to the second object, and allocating the budget with the third difference value to the contact and/or the contact group which meets the preset condition, wherein the contact and/or the contact group which is newly added with the budget with the third difference value still meets the respective constraint condition;
if the first value is less than the second value and the difference is a fourth difference, reducing the budget for the fourth difference from the expenditure of the contact and/or the contact group for which the constraint has been met, and allocating the budget for the fourth difference and the budget for the first value to the second object, wherein the contact and/or the contact group for which the budget for the fourth difference has been reduced still meet the respective constraint.
7. The method of claim 6, wherein:
when the first object is a contact group, the budget with the size of the first value is obtained by the following steps:
acquiring income information of each contact in the contact group;
selecting one or at least two contacts from the contact group according to the sequence of the profit information of each contact in the contact group from low to high, and executing budget reduction operation with the total number of the budget reduction operation being a first value;
and/or the presence of a gas in the atmosphere,
when the second object is a contact group, the budget having the second value is obtained by:
acquiring income information of each contact in the contact group;
and selecting one or at least two contact points from the contact point group according to the sequence from high to low of the profit information of each contact point in the contact point group, and executing the budget increase operation with the total number of the budget increase operation as a second value.
8. An advertisement budget allocation apparatus, comprising:
the determining module is used for determining the contact which does not meet the constraint condition and/or the contact group consisting of at least two contacts if the solving result of the preset budget allocation model does not meet the constraint condition of the budget allocation model, so as to obtain a target object;
the adjusting module is used for adjusting the budget of the target object, wherein the target object after the budget is adjusted meets the constraint condition;
the updating module is used for updating the solving result according to the adjusted budget;
and the output module is used for outputting the updated solving result.
9. An electronic device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, wherein the processor implements the method of any one of claims 1 to 7 when executing the computer program.
10. A computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the method according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210249490.3A CN114693343A (en) | 2022-03-14 | 2022-03-14 | Advertisement budget allocation method, device, equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210249490.3A CN114693343A (en) | 2022-03-14 | 2022-03-14 | Advertisement budget allocation method, device, equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114693343A true CN114693343A (en) | 2022-07-01 |
Family
ID=82140007
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210249490.3A Pending CN114693343A (en) | 2022-03-14 | 2022-03-14 | Advertisement budget allocation method, device, equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114693343A (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106961125A (en) * | 2017-05-08 | 2017-07-18 | 西安理工大学 | The equality constraint processing method of wind Thermal and Hydroelectric Power Systems dynamic economic dispatch |
| CN111882440A (en) * | 2020-05-27 | 2020-11-03 | 马上消费金融股份有限公司 | Asset securitization fund pool optimization method and device and storage medium |
| CN113191830A (en) * | 2021-07-02 | 2021-07-30 | 北京明略软件系统有限公司 | Resource allocation method, device, equipment and computer readable medium |
| CN113393269A (en) * | 2021-06-11 | 2021-09-14 | 上海明略人工智能(集团)有限公司 | Method and device for determining conversion rate of contact medium, electronic equipment and storage medium |
| CN113690897A (en) * | 2021-08-04 | 2021-11-23 | 国电南瑞科技股份有限公司 | Method and system for online dynamic optimization adjustment of low-frequency load shedding control objects in each turn |
-
2022
- 2022-03-14 CN CN202210249490.3A patent/CN114693343A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106961125A (en) * | 2017-05-08 | 2017-07-18 | 西安理工大学 | The equality constraint processing method of wind Thermal and Hydroelectric Power Systems dynamic economic dispatch |
| CN111882440A (en) * | 2020-05-27 | 2020-11-03 | 马上消费金融股份有限公司 | Asset securitization fund pool optimization method and device and storage medium |
| CN113393269A (en) * | 2021-06-11 | 2021-09-14 | 上海明略人工智能(集团)有限公司 | Method and device for determining conversion rate of contact medium, electronic equipment and storage medium |
| CN113191830A (en) * | 2021-07-02 | 2021-07-30 | 北京明略软件系统有限公司 | Resource allocation method, device, equipment and computer readable medium |
| CN113690897A (en) * | 2021-08-04 | 2021-11-23 | 国电南瑞科技股份有限公司 | Method and system for online dynamic optimization adjustment of low-frequency load shedding control objects in each turn |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110033314B (en) | Advertisement data processing method and device | |
| CN108364190B (en) | Mobile crowd sensing online excitation method combined with reputation updating | |
| Raglend et al. | Solution to profit based unit commitment problem using particle swarm optimization | |
| JP6876048B2 (en) | Predictive segmentation of energy customers | |
| CN109636482B (en) | Data processing method and system based on similarity model | |
| CN106910091A (en) | Data processing method and device | |
| CN113159835B (en) | Power generation side electricity price quotation method and device based on artificial intelligence, storage medium and electronic equipment | |
| CN111967971B (en) | Bank customer data processing method and device | |
| CN112183818A (en) | Recommendation probability prediction method and device, electronic equipment and storage medium | |
| CN110969490A (en) | Advertisement putting method and device | |
| CN112163886B (en) | Real-time bidding advertisement resource allocation method based on reinforcement learning | |
| CN103631939A (en) | Data processing method and data processing device for search engine | |
| CN108364198A (en) | A kind of online motivational techniques of mobile crowdsourcing based on social networks | |
| US20220327495A1 (en) | Intelligent scheduling using a prediction model | |
| Flajolet et al. | Real-time bidding with side information | |
| CN111814062A (en) | Information pushing method and device, server and storage medium | |
| CN110533437B (en) | Advertisement delivery budget allocation method and device | |
| Du et al. | Adversarial deep learning for online resource allocation | |
| CN115730125A (en) | Object identification method and device, computer equipment and storage medium | |
| CN114693343A (en) | Advertisement budget allocation method, device, equipment and storage medium | |
| CN110069708B (en) | Cross-medium popularization promotion effect estimation method, device, medium and equipment | |
| CN106557871A (en) | A kind of method for allocating tasks in gunz system based on stable matching algorithm | |
| CN110215703A (en) | The selection method of game application, apparatus and system | |
| CN106657058B (en) | Event resource allocation method and device | |
| CN115185606A (en) | Method, device, equipment and storage medium for obtaining service configuration parameters |
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 |