CN116976998A - Method and device for determining bid strategy of effect advertisement putting - Google Patents

Method and device for determining bid strategy of effect advertisement putting Download PDF

Info

Publication number
CN116976998A
CN116976998A CN202210398773.4A CN202210398773A CN116976998A CN 116976998 A CN116976998 A CN 116976998A CN 202210398773 A CN202210398773 A CN 202210398773A CN 116976998 A CN116976998 A CN 116976998A
Authority
CN
China
Prior art keywords
cost
business
term
preset
determining
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
Application number
CN202210398773.4A
Other languages
Chinese (zh)
Inventor
陈思宇
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Beta Technology Beijing Co ltd
Original Assignee
Beta Technology Beijing Co ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Beta Technology Beijing Co ltd filed Critical Beta Technology Beijing Co ltd
Priority to CN202210398773.4A priority Critical patent/CN116976998A/en
Publication of CN116976998A publication Critical patent/CN116976998A/en
Pending legal-status Critical Current

Links

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
    • G06Q30/0273—Determination of fees for advertising
    • 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
    • G06Q30/0277—Online advertisement

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 provides a method and a device for determining an effect advertisement putting bid strategy, comprising the following steps: determining a preset target business opportunity cost; determining the actual business opportunity cost in a preset time period; obtaining a business opportunity cost error item according to the target business opportunity cost and the actual business opportunity cost; obtaining a cost price adjustment coefficient according to the business opportunity cost error item; and determining an effect advertisement putting bidding strategy based on the business opportunity cost according to the target business opportunity cost, the preset exposal probability and the cost price adjustment coefficient. The accumulated situation of the differences of the business machine costs in the time period between the updated time points is obtained by acquiring the actual business machine costs in the preset time period instead of the updated time points, and the business machine cost differences in a period of time are more fully considered, so that the effect advertisement putting situation is reflected more accurately, and more accurate bidding is calculated.

Description

Method and device for determining bid strategy of effect advertisement putting
Technical Field
The invention relates to the technical field of Internet, in particular to an effect advertisement putting bidding method and device.
Background
The cost of advertising effectiveness is calculated by ocpm (Optimized Cost per Mille), which is more common in the prior art. In the ocpm-mode effective advertisement bidding calculation, PID calculation is performed on the cost difference between the target click cost and the actual click cost as a variable, thereby further obtaining the effective advertisement bidding. Since only the PID calculation of the cost difference between the target click cost and the actual click cost at each observed time point is considered in the PID calculation of the cost difference, the cumulative situation of the cost difference between the target click cost and the actual click cost in the time period between the updated observed time points is not considered, and thus, the bid value calculated thereby may be inaccurate.
Disclosure of Invention
The invention provides a method and a device for determining an effect advertisement putting bid strategy, which are used for solving the defect of inaccurate bid values calculated by an ocpm mode in the prior art and increasing the accuracy of bid calculation by improving the bid calculation mode.
The invention provides a method for determining an effect advertisement putting bid strategy, which comprises the following steps:
determining a preset target business opportunity cost;
determining the actual business opportunity cost in a preset time period;
obtaining a business opportunity cost error item according to the target business opportunity cost and the actual business opportunity cost;
Obtaining a cost price adjustment coefficient according to the business opportunity cost error item;
and determining an effect advertisement putting bidding strategy based on the business opportunity cost according to the target business opportunity cost, the preset exposal probability and the cost price adjustment coefficient.
According to the method for determining the bid strategy of the advertising impression, which is provided by the invention, the actual business cost in the preset time period is determined, and the method specifically comprises the following steps:
obtaining the total number of business opportunities in the preset time period according to the actual number of business opportunities in the preset time period and the number of behaviors which are not converted into business opportunities and are in semi-conversion in the preset time period; the semi-conversion behavior data in a preset time period is input into a semi-conversion behavior business model to obtain the behavior number which is generated in the semi-conversion behavior and is not converted into business opportunities, wherein the semi-conversion behavior is preset key behavior in the process of converting the put effect advertisement into the business opportunities;
and determining the actual business opportunity cost according to the total business opportunity number and the actual consumed cost of the effect advertisement delivery in the preset time period.
According to the method for determining the bid strategy of the advertising impression, which is provided by the invention, the actual business cost in the preset time period is determined, and the method comprises the following steps:
Determining whether the effect advertisement is in a cold start operation stage at the target moment according to the throwing condition of the effect advertisement and a preset judging standard; the cold start operation stage is a release stage with sparse operation data.
According to the method for determining the bid strategy of the advertising effect provided by the invention, the actual business cost in the preset time period is determined in response to the advertising effect being in the non-cold start operation stage, and the method specifically comprises the following steps:
determining a first short-term actual business cost, wherein the first short-term actual business cost is the business cost actually occurring in a first preset short-term period of time before a target moment;
determining a first mid-term actual business cost, wherein the first mid-term actual business cost is the business cost actually occurring in a first preset mid-term period of time before a target moment;
determining a first long-term actual business cost, wherein the first long-term actual business cost is a business cost actually occurring in a first preset long-term period of time before a target moment;
the duration of the first preset short-term time period is smaller than the duration of the first preset middle-term time period, and the duration of the first preset middle-term time period is smaller than the duration of the first preset long-term time period.
According to the method for determining the bid strategy of the advertising effect provided by the invention, the actual business cost in the preset time period is determined in response to the advertising effect in the cold start operation stage, and the method specifically comprises the following steps:
determining a second short-term actual business cost, wherein the second short-term actual business cost is the business cost actually occurring in a second preset short-term period of time before a target moment;
determining a second mid-term actual business cost, wherein the second mid-term actual business cost is the business cost actually occurring in a second preset mid-term period of time before a target moment;
determining a second long-term actual business cost, wherein the second long-term actual business cost is a business cost actually occurring within a second preset long-term period of time before a target moment; the duration of the second preset short-term time period is smaller than the duration of the second preset middle-term time period, and the duration of the second preset middle-term time period is smaller than the duration of the second preset long-term time period;
and taking the largest value of the second short-term actual business opportunity cost, the second medium-term actual business opportunity cost and the second long-term actual business opportunity cost as the actual business opportunity cost of the cold start operation stage.
According to the method for determining the bid strategy of the advertising impression of the effect provided by the invention, the business cost error item is obtained according to the target business cost and the actual business cost, and the method specifically comprises the following steps:
obtaining a short-term business opportunity cost error according to the difference value between the target business opportunity cost and the first short-term actual business opportunity cost;
obtaining a middle-term business opportunity cost error according to the difference value between the target business opportunity cost and the first middle-term actual business opportunity cost;
obtaining a long-term business opportunity cost error according to the difference value between the target business opportunity cost and the first long-term actual business opportunity cost;
determining a preset first proportional coefficient, an integral coefficient and a differential coefficient;
and obtaining a business opportunity cost error item in a non-cold start operation stage according to the short-term business opportunity cost error, the medium-term business opportunity cost error, the long-term business opportunity cost error and the differential coefficient, the first proportional coefficient and the integral coefficient which are respectively and sequentially corresponding.
According to the method for determining the bid strategy of the advertising impression of the effect provided by the invention, the business cost error item is obtained according to the target business cost and the actual business cost, and the method specifically comprises the following steps:
obtaining a business opportunity cost error of the cold starting operation stage according to the difference value between the target business opportunity cost and the actual business opportunity cost of the cold starting operation stage;
And obtaining a business opportunity cost error item of the cold start operation stage according to the business opportunity cost error of the cold start operation stage and a preset second proportion coefficient.
According to the method for determining the bid strategy of the advertising effect provided by the invention, the cost price adjustment coefficient is obtained according to the business machine cost error item, and the method specifically comprises the following steps:
the cost price adjustment coefficient is expressed as:
wherein k is t For the cost price adjustment coefficient at time t, k t-1 Is the cost price adjustment coefficient at the time (t-1),u is based on e t Power of th, u t Is a business opportunity cost error term;
when the effect advertisement is put in a non-cold start operation stage:
wherein u is t1 Commercial opportunity cost error term, k for effect advertising during non-cold start-up run phase p1 As a first scale factor, k i Is an integral coefficient, k d As a result of the differential coefficient,is the cost error of medium-term business opportunity，/>For long term business cost error->CPA for short term business cost error target CPA for the purpose of cost of business opportunity a CPA is the first short term actual business cost b CPA is the first medium-term actual business cost c Is the first long-term actual business cost;
when the effect advertisement is put in the cold start operation stage:
u t2 ＝k p2 *E t2
E t2 ＝CPA target -max(CPA d ，CPA e ，CPA f )
wherein u is t2 Commercial opportunity cost error term, k of effect advertisement in cold start operation stage p2 For a second proportionality coefficient, E t2 CPA for cold start operation phase business opportunity cost error d CPA for the second short-term actual business cost e CPA is the second medium-term actual business cost f Is the second long-term actual business cost.
According to the method for determining the effect advertisement putting bid strategy provided by the invention, the training process of the semi-conversion behavior business model specifically comprises the following steps:
determining a sample of which the half-conversion behavior occurs in the time range of the preset attribution time window, setting the sample of which the half-conversion behavior is converted into the business opportunity in the time range of the preset attribution time window as a positive sample, and setting the sample of which the half-conversion behavior is not converted into the business opportunity in the time range of the preset attribution time window as a negative sample;
acquiring characteristic data of preset attributes of the sample;
determining the type of a training model, training the training model through the sample and the characteristic data of the sample, and taking the trained training model as a half-conversion behavior business model.
The invention also provides a device for determining the bid strategy of the effect advertisement delivery, which comprises the following steps:
the target business machine cost determining unit is used for determining preset target business machine cost;
the actual business opportunity cost determining unit is used for determining the actual business opportunity cost in a preset time period;
The business opportunity cost error item acquisition unit is used for acquiring a business opportunity cost error item according to the target business opportunity cost and the actual business opportunity cost;
the cost price adjustment coefficient acquisition unit is used for acquiring a cost price adjustment coefficient according to the business opportunity cost error item;
and the effect advertisement putting bid strategy determining unit is used for determining the effect advertisement putting bid strategy based on the business opportunity cost according to the target business opportunity cost, the preset exposal business probability and the cost price adjustment coefficient.
According to the method and the device for determining the effect advertisement putting bid strategy, the business cost error item in the preset time period is obtained by obtaining the actual business cost in the preset time period, the cost price adjustment coefficient is adjusted through the business cost error item, so that the effect advertisement putting bid strategy is determined through the adjusted cost price adjustment coefficient, the effect advertisement putting bid is determined through the accumulated business cost error item in the preset time period, the effect advertisement putting situation in a period is reflected more accurately, the effect advertisement putting situation in each time point is not reflected, and the effect advertisement putting bid is obtained more accurately and objectively.
Drawings
In order to more clearly illustrate the invention or the technical solutions of the prior art, the following description will briefly explain the drawings used in the embodiments or the description of the prior art, and it is obvious that the drawings in the following description are some embodiments of the invention, and other drawings can be obtained according to the drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flow chart of an effect advertising bidding method provided by the present invention;
FIG. 2 is a schematic diagram of the relationship between the effectiveness advertisement semi-conversion behavior and the generator business;
FIG. 3 is a flow chart of a method of the corresponding semi-transformation behavior business model training process of step 120 of FIG. 1;
FIG. 4 is a flowchart of a specific method for providing an advertisement for an effect during a non-cold start operation according to the present invention, corresponding to step 130 in FIG. 1;
FIG. 5 is a flowchart of a specific method for providing an advertisement for an effect corresponding to step 130 of FIG. 1 during a cold start operation phase according to the present invention;
FIG. 6 is a schematic diagram of an effective advertising bidding device of the present invention;
fig. 7 is a schematic structural diagram of an electronic device provided by the present invention.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the present invention more apparent, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings, and it is apparent that the described embodiments are some embodiments of the present invention, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
In the prior ocpm control cost technology, the bidding formula is as follows:
ocpm(bid t )＝ctr*cpc target *k t (1)
wherein bid t For bidding, ocpm (bid) t ) Bidding when controlling cost for ocpm, ctr is the conversion rate of exposure to click action in preset effect advertisement delivery, cpc target To target click cost, k t And (5) adjusting the price coefficient for the cost at the time t.
The cost price adjustment coefficient k t In the prior art, a PID regulation mode is adopted.
e(t)＝cpc target (t)-cpc real (t) (2)
Wherein, the one-day delivery process can be divided into 24 parts, and each hour is taken as a time point (i.e. t). By presetting an update time interval pair k t Performing update calculations, i.e. k is updated every hour t 。
In the formula (2), e (t) is the cost difference at the time t of the observed time point cpc target (t) is the target click cost, cpc, at the time point t of observation real (t) is the actual click cost at the time point t of observation, k in formula (3) p ，k i ，k d The hyper-parameters searched for the grid search method correspond to proportional term coefficients, integral term coefficients, and differential term coefficients in the PID algorithm. In the formula (4)U is based on e t Power of the power of U t When greater than 0, it is mapped to a number greater than 1, whereas when u t When the number is smaller than 0, the number is mapped into a number smaller than 1, so that k is realized t The whole process can be understood as reducing k when the cost is higher than expected (target cost) t-1 Thereby reducing the value of the valence coefficient k at time t t Further influencing the bid price of the reference at the time t, otherwise increasing k when the cost is lower than expected t-1 To increase the value of the price adjustment coefficient k at the time t t And further influences the bid and bid of the reference at the time t, thereby realizing the ocpm bid of the effect advertisement putting by taking the click cost as the target in the effect advertisement putting process.
The ocpm bid calculation method mainly has the following problems: (1) The PID error calculation mode is used for calculating the error of each updated single point moment, and even if an integral term exists, only the accumulation of the error of each time point is considered, and the influence caused by the flow of each time point is not considered; (2) During the cold start run phase of an effect advertisement, the data may be untrusted and the cost of a newly placed effect advertisement may not be effectively controlled using a uniform bid; (3) The existing cost control problem only aims at the click behavior which can be fed back in time, but an effect advertiser has the requirement of controlling the cost of deeper transformation behavior, but in the existing scene, the control of the business opportunity cost is difficult, because the attribution window of the business opportunity is very long, and the business opportunity behavior cannot be recovered in time.
As shown in FIG. 1, an embodiment of the present invention provides a method for determining an effective advertisement delivery bidding strategy, comprising:
step 110: determining a preset target business opportunity cost; .
Specifically, the target business opportunity cost is business opportunity cost preset for the put effect advertisement, and the meaning of the business opportunity in the embodiment of the invention is as follows: the put effect advertisement is converted into a behavior of a preset conversion effect.
Step 120: and determining the actual business opportunity cost in the preset time period.
In the embodiment of the present invention, the determining the actual opportunity cost in the preset time period corresponding to step 120 specifically includes:
obtaining the total number of business opportunities in the preset time period according to the actual number of business opportunities in the preset time period and the number of behaviors which are not converted into business opportunities and are in semi-conversion in the preset time period; the semi-conversion behavior data in a preset time period is input into a semi-conversion behavior business model to obtain the behavior number which is generated in the semi-conversion behavior and is not converted into business opportunities, wherein the semi-conversion behavior is preset key behavior in the process of converting the put effect advertisement into the business opportunities;
and determining the actual business opportunity cost according to the total business opportunity number and the actual consumed cost of the effect advertisement delivery in the preset time period.
In the embodiment of the invention, the actual business cost is expressed as:
wherein CPA is real For the actual business cost in the time period (t-n) to t, cost is the cost actually consumed for effective advertisement delivery in the time period (t-n) to t, and the com is the total number of business,is the total number of business opportunities actually occurring in the period from (t-n) to t, +.>Is the total number of behaviors which are half-converted and not converted into business in the period from (t-n) to (t)>The method is obtained by inputting a half-conversion behavior business model through half-conversion behaviors occurring in a time period from (t-n) to (t), wherein the half-conversion behavior is a key behavior in the process of converting the put effect advertisement into business.
In the embodiment of the invention, a semi-conversion behavior business model is introduced, and the effect advertisement in the real estate field shown in fig. 2 is taken as an example for a scene with a longer time window for converting the put effect advertisement into a business opportunity. Because the time window of the conversion behavior is very long, the conversion behavior cannot be timely attributed to the subsequent business behavior after the effect advertisement is exposed, but the semi-conversion behavior of the user can be timely observed, and key behaviors such as clicking, calling, page visit, sharing and collection of the user are timely observed, in order to calculate the actual business cost of the effect advertisement more timely and accurately, a semi-conversion behavior business model is trained, and CPA caused by too low statistical value of the total business is relieved real Overestimating the problem, thereby effectively controlling the delivery cost of the dsp advertisements.
Specifically, as shown in fig. 3, the training process of the semi-transformation behavior business model specifically includes:
step 310: and determining samples of which the half-conversion behavior occurs in the time range of the preset attribution time window, setting the samples of which the half-conversion behavior is converted into the business opportunity in the time range of the preset attribution time window as positive samples, and setting the samples of which the half-conversion behavior is not converted into the business opportunity in the time range of the preset attribution time window as negative samples.
Specifically, the sample with half conversion behavior (clicking, calling, page visit, sharing, collection, etc. of the user) is screened, the preset attribution time window is selected for seven days, that is, the sample with half conversion behavior in seven days, the label (label) of the sample with business opportunity in the sample is set to be 1 as a positive sample by using the last attribution method, otherwise, the label (label) is 0 as a negative sample.
Step 320: and acquiring characteristic data of preset attributes of the sample.
Specifically, for example, the activity of the user in the sample in the station is counted for approximately 1, 3, 7, 30 and 60 days, such as the activity times, clicking times, call-up times, page visit times, sharing times, collection times and business times of the user, the exposure times, clicking times of the user in the dsp putting platform, the clicking rate, call-up rate and business probability of the effect advertisement position, and the like.
Step 330; determining the type of a training model, training the training model through the sample and the characteristic data of the sample, and taking the trained training model as a half-conversion behavior business model.
Specifically, for example, an FM model is selected as a training model, which is a very basic and important model in the field of recommendation and advertising because it can still have excellent performance and effect in a data-scarce scene. For each feature x i All make corresponding hidden vector of k dimensionAnd let->Thus, the calculation formula of the FM model can be obtained: />As long as there is a sum X in the data set X j Features that occur simultaneously, in which case v can be learned j In this case even in training set x i And x j Not at the same time, but due to v i And v j Appears in other combinations of features and thus their representation can still be learned. When x appears in the test set i x j In the case of =1, it is also possible to pass +.>To calculate the importance of the interaction of these two features.
In the embodiment of the present invention, the determining the actual business opportunity cost in the preset time period corresponding to step 120 further includes:
determining whether the effect advertisement is in a cold start operation stage at the target moment according to the throwing condition of the effect advertisement and a preset judging standard; the cold start operation stage is a release stage with sparse operation data.
In the embodiment of the invention, when the effect advertisement is put in the cold start operation stage, the data is sparse and the data is not confidence, so that in order to effectively control the cost, the stage of the effect advertisement is required to be distinguished from the normal operation stage (namely, the non-cold start operation stage) after the cold start operation stage, so that the specific cost control is carried out for the specific operation stage, and the accurate cost control is realized. Specifically, the preset determination criteria include, but are not limited to, determining whether the effect advertisement is in a cold start operation stage based on the time of putting the effect advertisement or the data of the effect advertisement.
Responding to the actual business cost in the determined preset time period in step 120 when the effect advertisement is in a non-cold start operation stage, that is, when the effect advertisement is in a normal operation stage after the cold start operation stage, specifically including:
and determining a first short-term actual business cost, wherein the first short-term actual business cost is the business cost actually occurring in a first preset short-term period before the target moment.
Specifically, in the normal phase of the effect ad operation, in order to observe the cumulative cost difference of the effect ad operation, the first short-term actual business cost, the first mid-term actual business cost, and the first long-term actual business cost are obtained to replace the differential term, the proportional term, and the integral term in the PID algorithm in ocpm. Specifically, CPA is used in the embodiment of the invention a Representing a first short-term actual business cost. For example, CPA takes the cost error 3 hours before the target time t as a short term trend of the cost of delivering the effect advertisement 3 hours The command is obtained according to the formula (5), and the command in the formula (6) according to the formula (5) is obtained according to the total number of business opportunities actually occurring within 3 hours from the target time t and the total number of behaviors which are semi-converted but not converted into business opportunities.
And determining a first middle-term actual business cost, wherein the first middle-term actual business cost is the business cost actually occurring in a first preset middle-term period before the target moment.
Specifically, CPA is used in the embodiment of the invention b Representing the first mid-term actual business costs. For example, CPA takes the cost error of 6 hours before the target time t as the mid-term trend of the cost of delivering the effect advertisement For 6 hours Reference to CPA for a specific calculation procedure 3 hours Is not described in detail herein.
A first long-term actual business cost is determined, wherein the first long-term actual business cost is a business cost that actually occurs within a first preset long-term period of time before a target time.
Specifically, CPA is used in the embodiment of the invention c Representing the first mid-term actual business costs. For example, CPA takes a cost error of 12 hours before the target time t as a short-term trend of the cost of delivering the effect advertisement For 12 hours Obtained according to equation (5), the specific calculation process refers to CPA 3 hours Is not described in detail herein.
In the embodiment of the invention, the duration of the first preset short-term period is smaller than the duration of the first preset middle-term period, and the duration of the first preset middle-term period is smaller than the duration of the first preset long-term period. Specifically, the time setting of the duration of the short-term period, the duration of the medium-term period, and the duration of the long-term period is flexibly set depending on the specific situation.
Responding to the effect advertisement being in the cold start operation stage, the determining the actual opportunity cost within the preset time period in the step 120 specifically includes:
and determining a second short-term actual business cost, wherein the second short-term actual business cost is the business cost actually occurring in a second preset short-term period of time before the target moment.
And determining second medium-term actual business costs, wherein the second medium-term actual business costs are business costs actually occurring in a second preset medium-term period of time before the target moment.
Determining a second long-term actual business cost, wherein the second long-term actual business cost is a business cost actually occurring within a second preset long-term period of time before a target moment; the duration of the second preset short-term period is smaller than the duration of the second preset middle-term period, and the duration of the second preset middle-term period is smaller than the duration of the second preset long-term period.
And taking the largest value of the second short-term actual business opportunity cost, the second medium-term actual business opportunity cost and the second long-term actual business opportunity cost as the actual business opportunity cost of the cold start operation stage.
Specifically, taking 0 time of day at the target time, 3 hours and 12 hours from the target time as three observation times of the cold start operation phase, the actual business opportunity cost of the cold start operation phase is max (CPA 3 hours ，CPA For 12 hours ，CPA When day 0 )。
Step 130: obtaining a business opportunity cost error item according to the target business opportunity cost and the actual business opportunity cost;
in response to the effect advertisement being in the non-cold start operation stage, the obtaining a business cost error term according to the target business cost and the actual business cost in step 130, as shown in fig. 4, specifically includes:
Step 410: and obtaining a short-term business opportunity cost error according to the difference value between the target business opportunity cost and the first short-term actual business opportunity cost.
Specifically, the short term business cost error is:
wherein, the liquid crystal display device comprises a liquid crystal display device,CPA for short term business cost error target CPA for the purpose of cost of business opportunity a For a first short term actual business cost.
Step 420: and obtaining a medium-term business opportunity cost error according to the difference value between the target business opportunity cost and the first medium-term actual business opportunity cost.
Specifically, the medium business cost error is:
wherein, the liquid crystal display device comprises a liquid crystal display device,CPA is the medium-term business opportunity cost error target CPA for the purpose of cost of business opportunity b Is the first medium-term actual business cost.
Step 430: and obtaining a long-term business opportunity cost error according to the difference value between the target business opportunity cost and the first long-term actual business opportunity cost.
Specifically, the long term business cost error is:
wherein, the liquid crystal display device comprises a liquid crystal display device,CPA for long term business cost error target CPA for the purpose of cost of business opportunity c Is the first long-term actual business cost.
Step 440: and determining a preset first proportional coefficient, an integral coefficient and a differential coefficient.
Specifically, the hyper-parameters searched by the grid search method are respectively used as the first scale coefficients k p1 Integral coefficient k i Differential coefficient k d 。
Step 450: and obtaining a business opportunity cost error item in a non-cold start operation stage according to the short-term business opportunity cost error, the medium-term business opportunity cost error, the long-term business opportunity cost error and the differential coefficient, the first proportional coefficient and the integral coefficient which are respectively and sequentially corresponding.
Specifically, the merchant cost error term u of the effect advertisement of normal stage operation t1 Expressed as:
in response to the effect advertisement being in the cold start operation stage, the obtaining a business cost error term according to the target business cost and the actual business cost in step 130, as shown in fig. 5, specifically includes:
step 510: and obtaining the business opportunity cost error of the cold start operation stage according to the difference value between the target business opportunity cost and the actual business opportunity cost of the cold start operation stage.
In particular, commercial cost error E of effect advertising at cold start run stage t2 Expressed as:
E t2 ＝CPA target -max(CPA d ，CPA e ，CPA f ) (11)
wherein CPA is d CPA for the second short-term actual business cost e CPA is the second medium-term actual business cost f Is the second long-term actual business cost.
Step 520: and obtaining a business opportunity cost error item of the cold start operation stage according to the business opportunity cost error of the cold start operation stage and a preset second proportion coefficient.
In particular, the commercial opportunity cost error term u for effect advertising at the cold start run stage t2 Expressed as:
u t2 ＝k p2 *E t2 (12)
wherein k is p2 Is the second scaling factor.
Step 140: obtaining a cost price adjustment coefficient according to the business opportunity cost error item;
in the embodiment of the present invention, the obtaining the cost price adjustment coefficient according to the business opportunity cost error term corresponding to step 140 specifically includes:
Cost price adjustment coefficient k t Expressed as:
wherein k is t For the cost price adjustment coefficient at time t, k t-1 Is the cost price adjustment coefficient at the time (t-1),u is based on e t Power of th, u t As a business cost error term, as can be seen from equation (13), when u t ＞0，k t With u t Is an increase to the power of e; in the embodiment of the invention, k t-1 Representing the corresponding k t Cost price adjustment coefficients at the previous updated point in time. As can be seen from the formula (13), for the effect advertisements in the cold start operation stage and the non-cold start operation stage, the corresponding business opportunity cost error items are obtained, and the corresponding cost price adjustment coefficients are obtained according to the same calculation mode.
Step 150: and determining an effect advertisement putting bidding strategy based on the business opportunity cost according to the target business opportunity cost, the preset exposal probability and the cost price adjustment coefficient.
In the embodiment of the present invention, the determining, according to the target business cost, the preset exposure business probability and the cost price adjustment coefficient in the corresponding step 150, an effect advertisement putting bidding strategy based on the business cost specifically includes:
bid ecpm (bid) of effect advertisement at target time t t ) The method comprises the following steps:
ecpm(bid t )＝pctdtr*CPA target *k t (14)。
specifically, in the embodiment of the present invention, the exposure business probability pctdtr is a known value predicted according to a known model.
As can be seen from equation (14), the bid-offer strategy for placement of the effect advertisement provided by the embodiments of the present invention is based on business cost control, wherein the cost price adjustment factor is obtained by considering the factors of the cumulative cost difference between the updated time points, the longer window time for attributing the effect advertisement, and the stage of operation of the effect advertisement.
The following describes an effect advertisement putting bidding device provided by the embodiment of the present invention, where the effect advertisement putting bidding device described below and the effect advertisement putting bidding method described above can be referred to correspondingly, as shown in fig. 6, the embodiment of the present invention provides an effect advertisement putting bidding device, including:
the target business cost determining unit 610 is configured to determine a preset target business cost.
The actual business opportunity cost acquiring unit 620 is configured to acquire an actual business opportunity cost in a preset period.
In the embodiment of the present invention, the actual business opportunity cost acquiring unit 620 includes:
the business opportunity total number determining subunit is used for obtaining the business opportunity total number in the preset time period according to the business opportunity number actually occurring in the preset time period and the behavior number which is half-converted in the preset time period and is not converted into the business opportunity; the semi-conversion behavior data in a preset time period is input into a semi-conversion behavior business model to obtain the behavior number which is generated in the semi-conversion behavior and is not converted into business opportunities, wherein the semi-conversion behavior is preset key behavior in the process of converting the put effect advertisement into the business opportunities.
And the business opportunity cost determination subunit is used for determining the actual business opportunity cost according to the total business opportunity number and the cost actually consumed by the effect advertisement delivery in the preset time period.
In the embodiment of the present invention, the total business opportunity number of the actual business opportunity cost obtaining unit 620 in determining the actual business opportunity cost is the sum of the total business opportunity number actually returned and the business opportunity result predicted by the semi-conversion behavior business opportunity model.
Wherein the actual business opportunity cost obtaining unit 620 includes a half-conversion behavior business opportunity model training subunit, and specifically the half-conversion behavior business opportunity model training subunit includes:
and the sample determination subunit is used for determining samples with half-conversion behaviors in the time range of the preset attribution time window, setting the samples with half-conversion behaviors converted into business opportunities in the time range of the preset attribution time window as positive samples, and setting the samples with half-conversion behaviors not converted into business opportunities in the time range of the preset attribution time window as negative samples.
A feature data obtaining subunit, configured to obtain feature data of a preset attribute of the sample; .
And the model training subunit is used for determining the type of the training model, training the training model through the sample and the characteristic data of the sample, and taking the trained training model as a semi-transformation behavior business model.
In the embodiment of the invention, the effect advertisement putting bidding device further comprises: the effect advertisement putting stage judging unit is used for determining whether the effect advertisement is in a cold start operation stage at the target moment according to the putting condition of the effect advertisement and a preset judging standard; the cold start operation stage is a release stage with sparse operation data.
Specifically, in response to the effect advertisement being in a non-cold start operation phase, the actual business cost acquisition unit 620 includes:
and the first short-term actual business cost determining subunit is used for determining the first short-term actual business cost, wherein the first short-term actual business cost is the business cost actually occurring in a first preset short-term period before the target moment.
And the first middle-term actual business cost determining subunit is used for determining the first middle-term actual business cost, wherein the first middle-term actual business cost is the business cost actually occurring in a first preset middle-term period before the target moment.
A first long-term actual business cost determination subunit, configured to determine a first long-term actual business cost, where the first long-term actual business cost is a business cost actually occurring in a first preset long-term period of time before a target time; the duration of the first preset short-term time period is smaller than the duration of the first preset middle-term time period, and the duration of the first preset middle-term time period is smaller than the duration of the first preset long-term time period.
In response to the effect advertisement being in the cold start operation phase, the actual business cost acquisition unit 620 includes:
and the second short-term actual business cost determining subunit is used for determining a second short-term actual business cost, wherein the second short-term actual business cost is the business cost actually occurring in a second preset short-term period before the target moment.
And the second middle-term actual business cost determining subunit is used for determining the second middle-term actual business cost, wherein the second middle-term actual business cost is the business cost actually occurring in a second preset middle-term period before the target moment.
A second long-term actual business cost determination subunit, configured to determine a second long-term actual business cost, where the second long-term actual business cost is a business cost actually occurring in a second preset long-term period of time before a target time; the duration of the second preset short-term period is smaller than the duration of the second preset middle-term period, and the duration of the second preset middle-term period is smaller than the duration of the second preset long-term period.
And the actual business cost determining subunit is used for taking the largest value of the second short-term actual business cost, the second medium-term actual business cost and the second long-term actual business cost as the actual business cost of the cold start operation stage.
And the business opportunity cost error item obtaining unit 630 is configured to obtain a business opportunity cost error item according to the target business opportunity cost and the actual business opportunity cost.
For effect advertisements during non-cold start operation, merchant cost error term acquisition unit 630 includes:
and the short-term business opportunity cost error determination subunit is used for obtaining the short-term business opportunity cost error according to the difference value between the target business opportunity cost and the first short-term actual business opportunity cost.
And the middle-term business machine cost error determination subunit is used for obtaining the middle-term business machine cost error according to the difference value between the target business machine cost and the first middle-term actual business machine cost.
And the long-term business machine cost error determination subunit is used for obtaining the long-term business machine cost error according to the difference value between the target business machine cost and the first long-term actual business machine cost.
And the first coefficient determination subunit is used for determining a preset first proportional coefficient, an integral coefficient and a differential coefficient.
And the business machine cost error item determining subunit is used for obtaining the business machine cost error item in the non-cold start operation stage according to the short-term business machine cost error, the medium-term business machine cost error, the long-term business machine cost error, and the differential coefficient, the first proportional coefficient and the integral coefficient which are respectively and sequentially corresponding.
For effect advertising at the cold start run stage, the merchant cost error term acquisition unit 630 includes:
and the business machine cost error determination subunit is used for obtaining the business machine cost error of the cold start operation stage according to the difference value between the target business machine cost and the actual business machine cost of the cold start operation stage.
And the business machine cost error item determining subunit is used for obtaining the business machine cost error item of the cold start operation stage according to the business machine cost error of the cold start operation stage and a preset second proportionality coefficient.
And a cost price adjustment coefficient obtaining unit 640, configured to obtain a cost price adjustment coefficient according to the business opportunity cost error term.
And the effect advertisement putting bid strategy determining unit 650 is configured to determine an effect advertisement putting bid strategy based on the business opportunity cost according to the target business opportunity cost, the preset exposer probability and the cost price adjustment coefficient.
In the embodiment of the present invention, the cost price adjustment coefficient obtaining unit 640 specifically includes:
the cost price adjustment coefficient is expressed as:
wherein k is t For the cost price adjustment coefficient at time t, k t-1 Is the cost price adjustment coefficient at the time (t-1),u is based on e t Power of th, u t Is a business opportunity cost error term;
When the effect advertisement is put in a non-cold start operation stage:
wherein u is t1 Commercial opportunity cost error term, k for effect advertising during non-cold start-up run phase p1 As a first scale factor, k i Is an integral coefficient, k d As a result of the differential coefficient,for mid-term business opportunity cost error, < > and->For long term business cost error->CPA for short term business cost error target CPA for the purpose of cost of business opportunity a CPA is the first short term actual business cost b CPA is the first medium-term actual business cost c Is the first long-term actual business cost;
when the effect advertisement is put in the cold start operation stage:
u t2 ＝k p2 *E t2 (12)
E t2 ＝CPA target -max(CPA d ，CPA e ，CPA f ) (11)
wherein u is t2 Commercial opportunity cost error term, k of effect advertisement in cold start operation stage p2 For a second proportionality coefficient, E t2 CPA for cold start operation phase business opportunity cost error d CPA for the second short-term actual business cost e CPA is the second medium-term actual business cost f Is the second long-term actual business cost.
An embodiment of the present invention provides an entity structure diagram of an electronic device, as shown in fig. 7, and the entity structure diagram of the electronic device may include: processor 710, communication interface (Communications Interface) 720, memory 730, and communication bus 740, wherein processor 710, communication interface 720, memory 730, and communication bus 740 complete communication among each other. Processor 710 may invoke logic instructions in memory 730 to perform an effectiveness advertising bidding method comprising: determining a preset target business opportunity cost; determining the actual business opportunity cost in a preset time period; obtaining a business opportunity cost error item according to the target business opportunity cost and the actual business opportunity cost; obtaining a cost price adjustment coefficient according to the business opportunity cost error item; and determining an effect advertisement putting bidding strategy based on the business opportunity cost according to the target business opportunity cost, the preset exposal probability and the cost price adjustment coefficient.
Further, the logic instructions in the memory 730 described above may be implemented in the form of software functional units and may be stored in a computer readable storage medium when sold or used as a stand alone product. Based on such understanding, the technical solution of the embodiments of the present invention may be embodied in essence or a part contributing to the prior art or a part of the technical solution, in the form of a software product stored in a storage medium, including several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a random access Memory (RAM, random Access Memory), a magnetic disk, or an optical disk, or other various media capable of storing program codes.
In another aspect, embodiments of the present invention also provide a computer program product comprising a computer program stored on a non-transitory computer readable storage medium, the computer program comprising program instructions which, when executed by a computer, enable the computer to perform the effect advertising bidding method provided by the methods described above, the method comprising: determining a preset target business opportunity cost; determining the actual business opportunity cost in a preset time period; obtaining a business opportunity cost error item according to the target business opportunity cost and the actual business opportunity cost; obtaining a cost price adjustment coefficient according to the business opportunity cost error item; and determining an effect advertisement putting bidding strategy based on the business opportunity cost according to the target business opportunity cost, the preset exposal probability and the cost price adjustment coefficient.
In yet another aspect, embodiments of the present invention further provide a non-transitory computer readable storage medium having stored thereon a computer program which, when executed by a processor, is implemented to perform the effect advertisement delivery bidding method provided by the methods described above: determining a preset target business opportunity cost; determining the actual business opportunity cost in a preset time period; obtaining a business opportunity cost error item according to the target business opportunity cost and the actual business opportunity cost; obtaining a cost price adjustment coefficient according to the business opportunity cost error item; and determining an effect advertisement putting bidding strategy based on the business opportunity cost according to the target business opportunity cost, the preset exposal probability and the cost price adjustment coefficient.
The apparatus embodiments described above are merely illustrative, wherein the elements illustrated as separate elements may or may not be physically separate, and the elements shown as elements may or may not be physical elements, may be located in one place, or may be distributed over a plurality of network elements. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art will understand and implement the present invention without undue burden.
From the above description of the embodiments, it will be apparent to those skilled in the art that the embodiments may be implemented by means of software plus necessary general hardware platforms, or of course may be implemented by means of hardware. Based on this understanding, the foregoing technical solution may be embodied essentially or in a part contributing to the prior art in the form of a software product, which may be stored in a computer readable storage medium, such as ROM/RAM, a magnetic disk, an optical disk, etc., including several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the method described in the respective embodiments or some parts of the embodiments.
Finally, it should be noted that: the above embodiments are only for illustrating the technical solution of the present invention, and are not limiting; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims (10)

1. An effectiveness advertising bidding strategy determination method, comprising:
determining a preset target business opportunity cost;
determining the actual business opportunity cost in a preset time period;
obtaining a business opportunity cost error item according to the target business opportunity cost and the actual business opportunity cost;
obtaining a cost price adjustment coefficient according to the business opportunity cost error item;
and determining an effect advertisement putting bidding strategy based on the business opportunity cost according to the target business opportunity cost, the preset exposal probability and the cost price adjustment coefficient.
2. The method for determining the bid strategy of advertising effectiveness according to claim 1, wherein the determining the actual opportunity cost within the preset time period specifically comprises:
obtaining the total number of business opportunities in the preset time period according to the actual number of business opportunities in the preset time period and the number of behaviors which are not converted into business opportunities and are in semi-conversion in the preset time period; the semi-conversion behavior data in a preset time period is input into a semi-conversion behavior business model to obtain the behavior number which is generated in the semi-conversion behavior and is not converted into business opportunities, wherein the semi-conversion behavior is preset key behavior in the process of converting the put effect advertisement into the business opportunities;
And determining the actual business opportunity cost according to the total business opportunity number and the actual consumed cost of the effect advertisement delivery in the preset time period.
3. The method of claim 2, wherein determining the actual business cost for the predetermined period of time further comprises:
determining whether the effect advertisement is in a cold start operation stage at the target moment according to the throwing condition of the effect advertisement and a preset judging standard; the cold start operation stage is a release stage with sparse operation data.
4. The method of claim 3, wherein determining the actual business cost for the preset time period in response to the effectiveness advertisement being in a non-cold start operation phase comprises:
determining a first short-term actual business cost, wherein the first short-term actual business cost is the business cost actually occurring in a first preset short-term period of time before a target moment;
determining a first mid-term actual business cost, wherein the first mid-term actual business cost is the business cost actually occurring in a first preset mid-term period of time before a target moment;
Determining a first long-term actual business cost, wherein the first long-term actual business cost is a business cost actually occurring in a first preset long-term period of time before a target moment;
the duration of the first preset short-term time period is smaller than the duration of the first preset middle-term time period, and the duration of the first preset middle-term time period is smaller than the duration of the first preset long-term time period.
5. The method for determining the bid strategy for advertising effectiveness according to claim 3 or 4, wherein the determining the actual business cost within the preset time period in response to the advertising effectiveness being in the cold start operation phase comprises:
determining a second short-term actual business cost, wherein the second short-term actual business cost is the business cost actually occurring in a second preset short-term period of time before a target moment;
determining a second mid-term actual business cost, wherein the second mid-term actual business cost is the business cost actually occurring in a second preset mid-term period of time before a target moment;
determining a second long-term actual business cost, wherein the second long-term actual business cost is a business cost actually occurring within a second preset long-term period of time before a target moment; the duration of the second preset short-term time period is smaller than the duration of the second preset middle-term time period, and the duration of the second preset middle-term time period is smaller than the duration of the second preset long-term time period;
And taking the largest value of the second short-term actual business opportunity cost, the second medium-term actual business opportunity cost and the second long-term actual business opportunity cost as the actual business opportunity cost of the cold start operation stage.
6. The method for determining an advertising bidding strategy based on the effect of claim 4, wherein obtaining a business cost error term according to the target business cost and the actual business cost specifically comprises:
obtaining a short-term business opportunity cost error according to the difference value between the target business opportunity cost and the first short-term actual business opportunity cost;
obtaining a middle-term business opportunity cost error according to the difference value between the target business opportunity cost and the first middle-term actual business opportunity cost;
obtaining a long-term business opportunity cost error according to the difference value between the target business opportunity cost and the first long-term actual business opportunity cost;
determining a preset first proportional coefficient, an integral coefficient and a differential coefficient;
and obtaining a business opportunity cost error item in a non-cold start operation stage according to the short-term business opportunity cost error, the medium-term business opportunity cost error, the long-term business opportunity cost error and the differential coefficient, the first proportional coefficient and the integral coefficient which are respectively and sequentially corresponding.
7. The method for determining an advertising bidding strategy for effect according to claim 5, wherein obtaining a business cost error term according to the target business cost and the actual business cost specifically comprises:
Obtaining a business opportunity cost error of the cold starting operation stage according to the difference value between the target business opportunity cost and the actual business opportunity cost of the cold starting operation stage;
and obtaining a business opportunity cost error item of the cold start operation stage according to the business opportunity cost error of the cold start operation stage and a preset second proportion coefficient.
8. The method for determining a bid strategy for advertising effectiveness according to claim 6 or 7, wherein the obtaining a cost price adjustment coefficient according to the business cost error term comprises:
the cost price adjustment coefficient is expressed as:
wherein k is t For the cost price adjustment coefficient at time t, k t-1 Is the cost price adjustment coefficient at the time (t-1),u is based on e t Power of th, u t Is a business opportunity cost error term;
when the effect advertisement is put in a non-cold start operation stage:
wherein u is t1 For non-cold start operationCommercial machine cost error term, k of line-stage effect advertisement p1 As a first scale factor, k i Is an integral coefficient, k d As a result of the differential coefficient,for mid-term business opportunity cost error, < > and->For long term business cost error->CPA for short term business cost error target CPA for the purpose of cost of business opportunity a CPA is the first short term actual business cost b CPA is the first medium-term actual business cost c Is the first long-term actual business cost;
when the effect advertisement is put in the cold start operation stage:
u t2 ＝k p2 *E t2
E t2 ＝CPA target -max(CPA d ，CPA e ，CPA f )
wherein u is t2 Commercial opportunity cost error term, k of effect advertisement in cold start operation stage p2 For a second proportionality coefficient, E t2 CPA for cold start operation phase business opportunity cost error d CPA for the second short-term actual business cost e CPA is the second medium-term actual business cost f Is the second long-term actual business cost.
9. The method for determining the bid strategy of advertising effectiveness according to claim 2, wherein the training process of the semi-transformation behavior business model specifically comprises:
determining a sample of which the half-conversion behavior occurs in the time range of the preset attribution time window, setting the sample of which the half-conversion behavior is converted into the business opportunity in the time range of the preset attribution time window as a positive sample, and setting the sample of which the half-conversion behavior is not converted into the business opportunity in the time range of the preset attribution time window as a negative sample;
acquiring characteristic data of preset attributes of the sample;
determining the type of a training model, training the training model through the sample and the characteristic data of the sample, and taking the trained training model as a half-conversion behavior business model.
10. An effectiveness advertising bidding strategy determination apparatus, comprising:
the target business machine cost determining unit is used for determining preset target business machine cost;
the actual business opportunity cost determining unit is used for determining the actual business opportunity cost in a preset time period;
the business opportunity cost error item acquisition unit is used for acquiring a business opportunity cost error item according to the target business opportunity cost and the actual business opportunity cost;
the cost price adjustment coefficient acquisition unit is used for acquiring a cost price adjustment coefficient according to the business opportunity cost error item;
and the effect advertisement putting bid strategy determining unit is used for determining the effect advertisement putting bid strategy based on the business opportunity cost according to the target business opportunity cost, the preset exposal business probability and the cost price adjustment coefficient.
CN202210398773.4A 2022-04-15 2022-04-15 Method and device for determining bid strategy of effect advertisement putting Pending CN116976998A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202210398773.4A CN116976998A (en) 2022-04-15 2022-04-15 Method and device for determining bid strategy of effect advertisement putting

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202210398773.4A CN116976998A (en) 2022-04-15 2022-04-15 Method and device for determining bid strategy of effect advertisement putting

Publications (1)

Publication Number Publication Date
CN116976998A true CN116976998A (en) 2023-10-31

Family

ID=88475351

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202210398773.4A Pending CN116976998A (en) 2022-04-15 2022-04-15 Method and device for determining bid strategy of effect advertisement putting

Country Status (1)

Country Link
CN (1) CN116976998A (en)

Similar Documents

Publication Publication Date Title
CN108269123B (en) Advertisement putting control method and device
Lu et al. A practical framework of conversion rate prediction for online display advertising
EP2757516A1 (en) System and method for serving electronic content
US20110231239A1 (en) Method and system for attributing an online conversion to multiple influencers
CN106296247B (en) Network information resource online ordering method and device
CN108965951B (en) Advertisement playing method and device
CN110880127B (en) Consumption level prediction method and device, electronic equipment and storage medium
CN109636477B (en) Early warning method and device for advertisement delivery shortage
CN110889725B (en) Online advertisement CTR estimation method, device, equipment and storage medium
CN111899041A (en) Information delivery processing method, information delivery device, information delivery equipment and storage medium
CN109615442A (en) RTB real time bid method based on excitation video ads
CN108932627B (en) Estimation method and device for consumption speed of CPA (continuous advertisement access) advertisement and readable storage medium
US20140058793A1 (en) Forecasting a number of impressions of a prospective advertisement listing
CN108564404B (en) Method and device for predicting return on investment of advertisement
US10181130B2 (en) Real-time updates to digital marketing forecast models
JP2019040529A (en) Correction device, correction method, and correction program
Austin et al. Reserve price optimization at scale
CN114862464A (en) Advertisement putting effect estimation method and device
WO2015051411A1 (en) Systems and methods for use in marketing
CN116976998A (en) Method and device for determining bid strategy of effect advertisement putting
CN110717653A (en) Risk identification method and device and electronic equipment
CN103942194B (en) A kind of information delivers the optimization method and device of account
CN114429362A (en) Advertisement product delivery method and device, electronic device and readable storage medium
CN112200389A (en) Data prediction method, device, equipment and storage medium
CN112446724A (en) Bidding method, device and equipment based on effect evaluation and readable storage medium

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