Advertisement conversion rate calibration method and device, electronic equipment and storage medium

Abstract

The application discloses an advertisement conversion rate calibration method and device, electronic equipment and a storage medium. The method comprises the following steps: obtaining a parameter cumulative value related to an estimated conversion rate of target advertisement data within a preset time length, each set of target advertisement data having the same dimension information under at least one dimension object and corresponding to the same identifier combination, and the target advertisement data being selected from M advertisement data; selecting a target parameter cumulative value from the parameter cumulative values of the multiple sets of target advertisement data corresponding to the preset time length according to the to-be-put advertisement data and the identifier combination corresponding to the to-be-put advertisement data; obtaining a calibration coefficient of the to-be-put advertisement data according to the target parameter cumulative value; and calibrating the estimated conversion rate of the to-be-put advertisement data by using the calibration coefficient to obtain a calibrated estimated conversion rate. By using the calibration method, the calibrated estimated conversion rate is more accurate.

Description

Technical Field

[0001] This application relates to the field of computer technology, and more specifically, to a method, apparatus, electronic device, and storage medium for calibrating advertising conversion rates. Background Technology

[0002] With the development of internet technology, online advertising has become the main way for many businesses to place ads. Advertisers or advertising platforms usually need to analyze the conversion rate of ads in order to calculate the cost of ads on the platform when placing ads.

[0003] Traditional conversion rate prediction methods typically employ either broad-based advertising or manual filtering of traffic to limit ad delivery. Only after accumulating a sufficient number of conversions are the ad display data trained to build a conversion rate prediction model for forecasting conversion rates. However, existing prediction methods do not yield accurate conversion rate estimates. Summary of the Invention

[0004] In view of this, embodiments of this application propose a method, apparatus, electronic device and storage medium for calibrating advertising conversion rates, which can calibrate the estimated conversion rate of advertising data to be delivered, so that the calibrated estimated conversion rate is more accurate.

[0005] In a first aspect, embodiments of this application provide a method for calibrating advertising conversion rates. The method includes: acquiring M advertising data and a corresponding identifier combination for each advertising data, wherein the M advertising data includes advertising data to be deployed, and each advertising data includes N dimension information, each dimension information corresponding to a different dimension object, where M is an integer greater than 1 and N is an integer greater than 1; acquiring cumulative parameter values ​​related to advertising conversion rates for multiple sets of target advertising data within a preset time period, wherein each set of target advertising data has the same dimension information and corresponds to the same identifier combination under at least one dimension object, and the target advertising data is selected from the M advertising data; selecting target parameter cumulative values ​​from the cumulative parameter values ​​corresponding to the multiple sets of target advertising data within the preset time period based on the advertising data to be deployed and the identifier combination corresponding to the advertising data to be deployed; obtaining a calibration coefficient for the advertising data to be deployed based on the target parameter cumulative values; and calibrating the estimated conversion rate of the advertising data to be deployed using the calibration coefficient to obtain a calibrated estimated conversion rate.

[0006] Secondly, embodiments of this application provide an advertising conversion rate calibration device, the device including a first acquisition module, a second acquisition module, a parameter cumulative value selection module, a calibration coefficient acquisition module, and a conversion rate calibration module. The first acquisition module is used to acquire M advertising data and the corresponding identifier combination for each advertising data. The M advertising data includes advertising data to be delivered. Each advertising data includes N dimension information, and each dimension information corresponds to a different dimension object. M is an integer greater than 1, and N is an integer greater than 1. The second acquisition module is used to acquire the cumulative parameter values ​​related to the advertising conversion rate of multiple sets of target advertising data within a preset time period. Each set of target advertising data has the same dimension information and corresponds to the same identifier combination under at least one dimension object. The target advertising data is selected from the M advertising data. The parameter cumulative value selection module is used to select the target parameter cumulative value from the cumulative parameter values ​​corresponding to the multiple sets of target advertising data within the preset time period based on the advertising data to be delivered and the identifier combination corresponding to the advertising data to be delivered. The calibration coefficient acquisition module is used to acquire the calibration coefficient of the advertising data to be delivered based on the target parameter cumulative value. The conversion rate calibration module is used to calibrate the estimated conversion rate of the advertising data to be delivered using the calibration coefficient to obtain the calibrated estimated conversion rate.

[0007] In one possible implementation, the parameter accumulation value selection module is further configured to select a target parameter accumulation value from the parameter accumulation values ​​corresponding to the multiple sets of target advertising data within a preset time period, based on the advertising data to be delivered, the identifier combination corresponding to the advertising data to be delivered, and the priority order of the N dimension objects.

[0008] In one possible implementation, the advertising consumption value range corresponding to each advertising data included in each group of target advertising data is the same, the preset duration includes a first preset duration and a second preset duration, and the first preset duration is less than the second preset duration. The second acquisition module is further used to acquire the cumulative value of a first parameter related to the advertising conversion rate of multiple groups of target advertising data within the first preset duration, and the cumulative value of a second parameter related to the advertising conversion rate of multiple groups of target advertising data within the second preset duration.

[0009] The parameter accumulation value selection module includes a first selection unit, a detection unit, and a second selection unit. The first selection unit is used to select, from the first parameter accumulation values ​​corresponding to the multiple sets of target advertising data, a first initial parameter accumulation value corresponding to each dimension of the advertising data to be delivered, wherein the identifier combination corresponding to the first initial parameter accumulation value is the identifier combination corresponding to the advertising data to be delivered. The detection unit is used to detect whether a first target consumption accumulation value exists among the first initial parameter accumulation values, wherein the minimum value of the advertising consumption range corresponding to the first target consumption accumulation value is greater than a preset advertising consumption threshold. The second selection unit is used, when no first target consumption accumulation value exists, to select a target parameter accumulation value from the second parameter accumulation values ​​corresponding to the multiple sets of target advertising data, according to the advertising data to be delivered, the identifier combination corresponding to the advertising data to be delivered, and the priority order of the N dimension objects.

[0010] In one possible implementation, the parameter accumulation value selection module further includes a quantity acquisition unit and a parameter determination unit. The quantity acquisition unit is used to acquire the quantity of a first target parameter accumulation value when a first target parameter accumulation value exists among the first initial parameter accumulation values. The parameter determination unit is used to select a target parameter accumulation value as the first target parameter accumulation value when there is only one first target parameter accumulation value; or to select a target parameter accumulation value from the at least two first target parameter accumulation values ​​based on the dimension information corresponding to the at least two first target parameter accumulation values, the dimension object to which the dimension information belongs, and the priority order of the N dimension objects when there are at least two first target parameter accumulation values.

[0011] In one possible implementation, the cumulative values ​​of the target parameters include the cumulative value of the target reported conversions, the cumulative value of the target estimated conversion rate, the cumulative value of the target estimated click-through rate, and the cumulative value of the target clicks. The calibration coefficient acquisition module includes: a first calibration coefficient acquisition unit, an exposure stage determination unit, a consumption stage determination unit, an expansion coefficient acquisition unit, and a second calibration coefficient acquisition unit. The first calibration coefficient acquisition unit is used to perform numerical calculations on the cumulative values ​​of the target reported conversions, the target estimated conversion rate, the target estimated click-through rate, and the target clicks to obtain the initial calibration coefficient of the advertising data to be deployed. The exposure stage determination unit is used to determine the exposure stage of the advertising data to be deployed based on the cumulative consumption value and the cumulative value of the reported conversions within the preset time period. The consumption stage determination unit is used to determine the consumption stage of the advertising data to be deployed based on the cumulative consumption value. The expansion coefficient acquisition unit is used to obtain the expansion coefficient based on the exposure stage and the consumption stage. The second calibration coefficient acquisition unit is used to process the initial calibration coefficient based on the expansion coefficient to obtain the calibration coefficient of the advertising data to be deployed.

[0012] In one possible implementation, the exposure stage determination unit is further configured to: determine the exposure stage of the advertising data to be delivered as a first exposure stage when the cumulative consumption value of the advertising data to be delivered within the preset duration is less than a preset cumulative consumption threshold, and the cumulative reported conversion value of the advertising data to be delivered within the preset duration is less than a preset conversion threshold; or determine the exposure stage of the advertising data to be delivered as a second exposure stage when the cumulative consumption value of the advertising data to be delivered within the preset duration is not less than a preset cumulative consumption threshold, or the cumulative reported conversion value of the advertising data to be delivered within the preset duration is not less than a preset conversion threshold.

[0013] In one possible implementation, the amplification coefficient acquisition unit is specifically used to: when the exposure stage is a first exposure stage and the consumption stage is a first consumption stage, acquire a preset constant, which is the amplification coefficient of the initial calibration coefficient; or when the exposure stage is a first exposure stage and the consumption stage is a second consumption stage, obtain a deviation value of the advertising data to be delivered based on the cumulative consumption value and cumulative conversion value of the advertising data to be delivered within the preset time period and the conversion bid of the advertising data to be delivered; obtain a first deviation correction parameter based on the cumulative consumption value of the advertising data to be delivered and a first target consumption value in the consumption range corresponding to the second consumption stage; and obtain an amplification coefficient based on the first deviation correction parameter and the deviation value, wherein the first... The maximum consumption value in the consumption range corresponding to the first consumption stage is less than the minimum consumption value in the consumption range corresponding to the second consumption stage; or, when the exposure stage is the first exposure stage and the consumption stage is the third consumption stage, the deviation value of the advertising data to be delivered is obtained based on the cumulative consumption value and cumulative conversion value of the advertising data to be delivered within the preset time period and the conversion bid of the advertising data to be delivered. A second deviation correction parameter is obtained based on the second target consumption value and the third consumption value in the consumption range corresponding to the third consumption stage, and an expansion coefficient is obtained based on the second deviation correction parameter and the deviation value, wherein the maximum consumption value in the consumption value range corresponding to the second consumption stage is less than the minimum consumption value in the consumption value range corresponding to the third consumption stage.

[0014] In one possible implementation, the amplification coefficient acquisition unit is further configured to, when the exposure stage is the second exposure stage, obtain a smoothing coefficient of the advertising data to be delivered based on the historical cumulative conversion value of the advertising data to be delivered and the preset duration; and obtain an amplification coefficient of the advertising data to be delivered based on the historical conversion volume and historical conversion rate estimated cumulative value of the advertising data to be delivered, the smoothing coefficient, and the target parameter cumulative value.

[0015] In one possible implementation, when there are multiple cumulative values ​​for the target parameters, the calibration device for the advertising conversion rate further includes a weighted calculation module for weighted summation of the multiple cumulative values ​​for the target parameters to obtain a weighted summation of the cumulative values ​​for the target parameters.

[0016] In one possible implementation, the conversion rate calibration module is further configured to multiply the calibration coefficient by the estimated conversion rate of the advertising data to be delivered, so as to obtain the calibrated estimated conversion rate.

[0017] In one possible implementation, the dimension object includes at least one of the advertiser dimension, product dimension, and advertiser dimension.

[0018] In one possible implementation, the identifier combination corresponding to the advertising data consists of the advertising publishing platform identifier and the advertising launch information identifier in the advertising data.

[0019] Thirdly, embodiments of this application provide an electronic device, including a processor and a memory; one or more programs are stored in the memory and configured to be executed by the processor to implement the above-described method.

[0020] Fourthly, embodiments of this application provide a computer-readable storage medium storing program code, wherein the above-described method is executed when the program code is run by a processor.

[0021] Fifthly, embodiments of this application provide a computer program product or computer program that includes computer instructions stored in a computer-readable storage medium. A processor of a computer device retrieves the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform the method described above.

[0022] This application provides a method, apparatus, electronic device, and storage medium for calibrating advertising conversion rates. During the calibration process for the estimated conversion rate of advertising data to be delivered, when selecting target parameter cumulative values ​​from multiple sets of target advertising data corresponding to parameter cumulative values ​​within a preset time period, the selected target parameter cumulative values ​​are the parameter cumulative values ​​corresponding to the target advertising data that have the same dimensional information and corresponding to the same dimensional combination as the advertising data to be delivered. That is, the target advertising data corresponding to the target parameter cumulative value is highly correlated with or similar to the advertising data to be delivered. Therefore, obtaining a calibration coefficient based on the target parameter cumulative value and using this calibration coefficient to calibrate the estimated conversion rate results in a more accurate calibrated estimated conversion rate. Attached Figure Description

[0023] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0024] Figure 1 A schematic diagram of a system architecture proposed in an embodiment of this application is shown;

[0025] Figure 2 A flowchart of an advertising conversion rate calibration method according to an embodiment of this application is shown;

[0026] Figure 3 This illustration shows the information included in each of the M advertising data presented in the embodiments of this application;

[0027] Figure 4 It shows from Figure 3 The selected advertising data has the same advertiser and the same identifier dimension as the advertising data to be published;

[0028] Figure 5 It shows from Figure 3 The selected advertising data has the same brand of advertised products and the same identification dimensions as the advertising data to be published.

[0029] Figure 6 It shows from Figure 3 The selected advertising data has the same advertising products and the same identification dimensions as the advertising data to be published;

[0030] Figure 7 Another flowchart of an advertising conversion rate calibration method proposed in an embodiment of this application is shown;

[0031] Figure 8 This illustration shows a schematic diagram of the data relationship proposed in an embodiment of this application;

[0032] Figure 9 Another flowchart of an advertising conversion rate calibration method proposed in an embodiment of this application is shown;

[0033] Figure 10 Another flowchart of an advertising conversion rate calibration method proposed in an embodiment of this application is shown;

[0034] Figure 11 The illustration shows the difference between the estimated conversion rate and the actual conversion rate of the advertising data to be delivered under the advertising product dimension provided in this application embodiment, obtained at different time periods.

[0035] Figure 12 This illustrates the distribution range of the ratio of the estimated conversion rate to the actual conversion rate before and after calibration, under the advertising product dimension provided in the embodiments of this application.

[0036] Figure 13 The illustration shows the difference between the estimated conversion rate and the actual conversion rate of the advertiser-to-be-delivered advertising data obtained before and after calibration at different time periods, based on the advertiser dimension provided in this application embodiment.

[0037] Figure 14 This illustrates the distribution range of the ratio of the estimated conversion rate to the actual conversion rate before and after calibration, under the advertiser dimension provided in the embodiments of this application.

[0038] Figure 15The diagram shows a connection block diagram of an advertising conversion rate calibration device according to an embodiment of this application;

[0039] Figure 16 It shows Figure 15 Connection diagram of the module for selecting cumulative values ​​of intermediate parameters;

[0040] Figure 17 It shows Figure 15 Connection block diagram of the calibration coefficient acquisition module;

[0041] Figure 18 A connection block diagram of an electronic device according to an embodiment of this application is shown. Detailed Implementation

[0042] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0043] The following describes the terms that may be involved in the embodiments of this application.

[0044] Advertising data typically includes dimensional information belonging to multiple dimensions, such as the advertiser to which the advertising data belongs, the specific brand that produces the advertising data, and the specific product that the advertising data corresponds to. Advertising data also includes identification information that can be used to identify the advertising platform and advertising launch information (such as the duration of the advertising launch).

[0045] Dimension objects are typically used to distinguish the data categories to which the various dimensional information included in advertising data belongs. The dimensional objects corresponding to advertising data include one or more of the following: advertiser dimension, advertised product brand dimension, and advertised product dimension. Different advertising data usually include different dimensional information. Therefore, each dimensional object can include multiple dimensional information.

[0046] Dimensional information specifically refers to the information included in each of the aforementioned dimensional objects. For example, when a dimensional object includes an advertiser dimension, the dimensional information included in that advertiser dimension is specifically advertiser information. For instance, if multiple advertising data include advertising data corresponding to advertisers A, B, and C, then the advertiser dimension includes the dimensional information for advertisers A, B, and C. As another example, when a dimensional object includes an advertised product brand dimension, the dimensional information included in that advertised product brand dimension is specifically product brand information. For instance, if multiple advertising data include advertising data corresponding to brands D and E, then the advertised product brand dimension includes the dimensional information for brands D and E. Furthermore, when a dimensional object includes an advertised product dimension, the dimensional information included in that advertised product dimension is specifically product information. For instance, if multiple advertising data include advertising data corresponding to products F and G, then the advertised product dimension includes the dimensional information for products F and G.

[0047] An identifier combination refers to an identifier combination consisting of at least two identifiers from advertising data. The identifiers in the advertising data include advertising platform identifiers, advertising launch information identifiers, and advertising audience identifiers. Specifically, the advertising platform identifier indicates which platform the advertising data is played on (e.g., platform I or platform J), the advertising launch information identifier indicates the duration of the advertisement's online presence (e.g., more than one day or one day; where more than one day corresponds to an old advertisement, and one day corresponds to a new advertisement), and the advertising audience identifier indicates which users typically view the advertising data. For example, if the identifier combination consists of an advertising platform identifier and an advertising launch information identifier, and the advertising platform identifier specifically includes platform I and platform J, and the advertising launch information identifier specifically includes new and old advertisements, then the identifier combination includes one or more of [platform I, new advertisement], [platform I, old advertisement], [platform J, new advertisement], and [platform J, old advertisement].

[0048] The cumulative parameter value refers to the summation of parameters obtained at different times within a preset time period. These parameters may include one or more of the following: ad conversions, estimated conversion rate, estimated click-through rate, number of clicks, and cost. Accordingly, the cumulative parameter value includes one or more of the following: cumulative conversion value, cumulative estimated conversion rate value, cumulative estimated click-through rate value, cumulative number of clicks, and cumulative cost.

[0049] Ad conversions refer to the number of times an ad is clicked and converted within a given time period. Correspondingly, the cumulative ad conversion count within a preset time period can be the total number of times an ad is clicked and converted within that preset time period.

[0050] Estimated conversion rate refers to the probability that an ad will convert after being clicked. Correspondingly, the cumulative estimated conversion rate is the sum of the estimated conversion rates obtained from clicking at multiple specified times within a preset time period.

[0051] Estimated click-through rate (CTR) refers to the probability that an ad will be clicked after it has been exposed. Correspondingly, the cumulative estimated CTR is the sum of the estimated CTRs obtained from exposing the ad at multiple specified times within a preset time period.

[0052] Click count refers to the number of times an ad is clicked within a given time period. Correspondingly, the cumulative click count is the total number of times an ad is clicked within a preset time period.

[0053] Advertising costs refer to the expenses incurred when an advertisement is played on a streaming platform. These costs can be calculated in several ways: First, by charging per thousand impressions (CPM). Second, by charging per click (CPC), where the advertiser pays the platform only if a user clicks on the ad. Third, by charging based on actual ad performance (CPA, Cost Per Action), such as charging based on valid responses to questionnaires or orders. Fourth, by charging based on comparative features (OCPA), where the billing point and bid point are separate; the billing point is at the click, and the bid point is at the conversion. In this application's implementation, the primary focus is on the CPA billing method, where the costs incurred by the advertisement on the streaming platform are calculated using CPA.

[0054] Specifically, for a given ad data point, if CPC (Cost Per Click) billing is used, the cost per click (Cost_per_click) = target_cpa * all_factor * PCVR, where targetCpa is the conversion bid, all_factor is a combination of industry factors, price adjustment factors, etc., and PCVR is the estimated conversion rate of the ad data. If CPM (Cost Per Mille) billing is used, the cost per exposure (Cost_per_exposure) = target_cpa * all_factor * PCVR * pctr, where PCTR is the estimated click-through rate of the ad data. As can be seen from the above formulas, the accuracy of the click-through rate estimation affects the fees charged by the platform, and similarly, it also affects the expenses incurred by advertisers.

[0055] The inventors discovered that advertising conversion rates are typically estimated using a conversion rate prediction model, based on the characteristics of the ad's information, user profile, and ad playback scenario within a historical timeframe. This model is trained on a large dataset of ad data, analyzing the conversion status (whether a conversion occurred) of each ad and the characteristics of that ad's information, user profile, and playback scenario within a preset timeframe. However, because the training set used is broad and lacks specificity, the predicted conversion rate for a particular ad is inaccurate. Therefore, inaccurate cost estimates based on conversion rate predictions can lead to losses for advertising platforms and advertisers.

[0056] In view of this, this application provides a method, apparatus, electronic device, and storage medium for calibrating advertising conversion rates. The method involves acquiring the cumulative parameter values ​​of multiple sets of target advertising data over a preset time period. Each set of target advertising data has the same dimensional information and corresponds to the same dimensional combination under at least one dimensional object. The target advertising data is selected from M sets of advertising data, which also include advertising data to be deployed. Based on the advertising data to be deployed and the target dimensional combination corresponding to it, cumulative target parameter values ​​are selected from the cumulative parameter values ​​of the multiple sets of target advertising data over the preset time period. A calibration coefficient for the advertising data to be deployed is obtained based on the cumulative target parameter values. This calibration coefficient is then used to calibrate the estimated conversion rate of the advertising data to be deployed, resulting in a calibrated estimated conversion rate. This method enables the acquisition of accurate calibration coefficients based on the cumulative target parameter values ​​of target advertising data that are highly correlated with or similar to the advertising data to be deployed during the calibration process. This makes the calibrated predicted conversion rate obtained by calibrating the predicted conversion rate using these calibration coefficients more accurate.

[0057] Specifically, the advertising conversion rate calibration method provided in this application selects target parameter cumulative values ​​from multiple sets of target advertising data corresponding to parameter cumulative values ​​within a preset time period. The selected target parameter cumulative values ​​are those corresponding to target advertising data that have the same dimensional information and corresponding to the same dimensional combination as the advertising data to be deployed. That is, the target advertising data corresponding to the target parameter cumulative value is highly correlated with or similar to the advertising data to be deployed. Therefore, obtaining a calibration coefficient based on the target parameter cumulative value and using this calibration coefficient to calibrate the estimated conversion rate results in a more accurate calibrated estimated conversion rate.

[0058] Figure 1 A schematic diagram of an exemplary system architecture 10 to which the technical solutions of the embodiments of this application can be applied is shown.

[0059] The system architecture 10 may include a terminal device 12, a server 11, and a network 13. The network 13 serves as the medium for providing a communication link between the server 11 and the terminal device 12. The network 13 may include various connection types, such as wired communication links, wireless communication links, etc.

[0060] It should be understood that Figure 1 The number of servers 11 and terminal devices 12 shown is merely illustrative. Depending on implementation needs, there can be any number of servers 11 and terminal devices 12.

[0061] The server 11 can be a standalone physical server 11, a server cluster 11 composed of multiple physical servers 11, or a distributed system. It can also be a cloud server 11 providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services 13, cloud communication, middleware services, domain name services, security services, content delivery network 13 (CDN), and big data and AI platforms. The terminal device 12 includes, but is not limited to, tablets, laptops, PDAs, mobile phones, voice interaction devices, and personal computers (PCs), but is not limited to these.

[0062] In one embodiment of this application, an advertiser can log in to server 11 to publish multiple advertising data through a content publishing platform associated with server 11. Users can then browse the advertising data published on the content publishing platform via terminal device 12, and can also click on the advertising data and purchase the corresponding products through the purchase links carried in the advertising data. It should be understood that different terminal devices 12 may display different advertising data. For example, the i-th terminal may display advertising data i1 and advertising data i2, and the j-th terminal may display advertising data j1 and advertising data j2.

[0063] This includes data on ads to be delivered. When calibrating the conversion rate of the data on ads to be delivered, the specific calibration process is as follows:

[0064] Server 11 acquires the cumulative parameter values ​​related to the estimated conversion rate of multiple sets of target ad data within a preset time period. Each set of target ad data has the same dimensional information and corresponds to the same identifier combination under at least one dimension object. The target ad data is selected from M ad data. Based on the ad data to be delivered and the identifier combination corresponding to it, the server selects the cumulative parameter values ​​of the multiple sets of target ad data within the preset time period. The server then obtains the calibration coefficient of the ad data to be delivered based on the cumulative target parameter values. Finally, the server uses the calibration coefficient to calibrate the estimated conversion rate of the ad data to be delivered, obtaining the calibrated estimated conversion rate. This allows the target ad data corresponding to the cumulative target parameter values ​​selected during the calibration process to be ad data that is highly correlated with or similar to the ad data to be delivered. Therefore, obtaining the calibration coefficient based on the cumulative target parameter values ​​and using the calibration coefficient to calibrate the estimated conversion rate results in a more accurate calibrated estimated conversion rate.

[0065] It should be noted that the advertising conversion rate calibration method provided in this application embodiment is generally executed by server 11, and correspondingly, the advertising conversion rate calibration device is generally set in server 11. However, in other embodiments of this application, terminal device 12 may also have similar functions to server 11, thereby executing the advertising conversion rate calibration method provided in this application embodiment.

[0066] It should be understood that after the server 11 completes the calibration of the estimated conversion rate of the advertising data to be delivered, it can also calculate the corresponding consumption based on the calibrated estimated conversion rate and deliver the advertising data to be delivered. The terminal device 12 can display the advertiser page of the advertising data to be delivered on its display interface. Users can click on the advertiser page to play the advertisement on the terminal device 12. Users can also perform corresponding conversion operations during the process of playing the advertisement on the terminal device 12 (such as when the advertisement to be delivered is a product sales advertisement, users can perform a purchase operation on the terminal device 12).

[0067] As an optional implementation, the server and terminal device in the above system architecture can also serve as nodes in the blockchain. The advertising data used in the advertising conversion rate calibration method disclosed in this application can be stored in the nodes, and the advertising conversion rate calibration method can be specifically executed by one or more nodes in the blockchain.

[0068] Blockchain is a new application model of computer technologies such as distributed data storage, peer-to-peer transmission, consensus mechanisms, and encryption algorithms. Essentially, a blockchain is a decentralized database, a chain of data blocks linked together using cryptographic methods. Each data block contains information about a batch of network transactions, used to verify the validity of the information (anti-counterfeiting) and to generate the next block.

[0069] The implementation details of the technical solutions in the embodiments of this application are described in detail below:

[0070] Figure 2 The flowchart illustrating an advertising conversion rate calibration method according to an embodiment of this application is shown in the schematic diagram. The entity executing the advertising conversion rate calibration method may be a server, for example, a... Figure 1 The server 104 shown can also be any terminal device with data processing capabilities.

[0071] Reference Figure 2 As shown, the method for calibrating the advertising conversion rate includes at least steps S110 to S150, which are detailed below:

[0072] Step S110: Obtain M advertising data and the corresponding identifier combination for each advertising data.

[0073] Among them, the M advertising data include advertising data to be delivered. Each advertising data includes N dimension information, and each dimension information corresponds to a different dimension object. M is an integer greater than 1, and N is an integer greater than 1.

[0074] The identifier combination can consist of at least two of the following: the ad playback platform identifier, the ad launch information identifier, and the ad audience identifier from the ad data.

[0075] The ad playback platform identifier indicates which specific ad platform the ad data is played on, such as platform I or platform J. The ad launch information identifier indicates whether the ad is a newly launched ad, such as a new ad or an old ad. An ad with a launch information identifier of "new ad" means that the ad's launch duration as of the current moment is within a preset duration threshold. An ad with a launch information identifier of "old ad" means that the ad's launch duration as of the current moment is greater than or equal to the preset duration threshold. The preset duration threshold can be one hour, one day, or one week, etc., and is not specifically limited here; it can be selected according to actual needs.

[0076] As one implementation method, the identifier combination corresponding to each advertising data includes an advertising playback platform identifier and an advertising launch information identifier for that advertising data.

[0077] In this approach, if the advertising platform identifiers in the M advertising data specifically include Platform I and Platform J, and the advertising launch information identifiers specifically include new ads and old ads, then the identifier combinations corresponding to the M advertising data include [Platform I, New Ad], [Platform I, Old Ad], [Platform J, New Ad], and [Platform J, Old Ad]. Correspondingly, the identifier combination for each of the M advertising data belongs to one of [Platform I, New Ad], [Platform I, Old Ad], [Platform J, New Ad], and [Platform J, Old Ad].

[0078] Each advertisement's data includes N dimensions, each belonging to a different dimension object; that is, there are N dimension objects. These include one or more of the following: advertiser dimension, advertised product brand dimension, and advertised product dimension.

[0079] As one implementation method, the dimension objects include advertiser dimension, advertised product brand dimension, and advertised product dimension.

[0080] The M advertising data and the corresponding identifier combination for each advertising data can be obtained from a database associated with the server or from the server itself; no specific limitation is made here.

[0081] Step S120: Obtain the cumulative values ​​of parameters related to the ad conversion rate for multiple sets of target ad data within a preset time period.

[0082] Each set of target ad data has the same dimensional information and the same identifier combination under at least one dimensional object. The target ad data is selected from M ad data.

[0083] The preset duration can include any one or more of the following: one hour, two hours, twelve hours, or one day.

[0084] In one optional implementation, the preset duration includes one day. That is, the step of obtaining the cumulative values ​​of parameters related to the ad conversion rate for multiple sets of target ad data within the preset duration includes: obtaining the cumulative values ​​of parameters related to the ad conversion rate for multiple sets of target ad data within one day.

[0085] The cumulative parameter value may include one or more of the following: cumulative conversion value, cumulative estimated conversion rate value, cumulative estimated click-through rate value, cumulative click value, and cumulative cost value.

[0086] In one implementation, the parameter cumulative values ​​include cumulative conversion count, cumulative estimated conversion rate, cumulative estimated click-through rate, cumulative click count, and cumulative cost.

[0087] It should be understood that each set of target ad data can include one or more ad data items. When a set of ad data includes multiple ad data items, the cumulative parameter value corresponding to that set of ad data can be obtained by summing the cumulative parameter values ​​of each ad data item included in that set.

[0088] Step S130: Based on the advertising data to be delivered and the corresponding identifier combination, select the target parameter cumulative value from the parameter cumulative values ​​corresponding to multiple sets of target advertising data within a preset time period.

[0089] There are several ways to select the cumulative value of the target parameter.

[0090] As one implementation, step S130 includes: selecting a set of target advertising data from multiple sets of target advertising data that has the same combination of markers as the advertising data to be delivered and at least one dimension information as the advertising data to be delivered, and using the cumulative value of the parameters corresponding to the set of target advertising data as the cumulative value of the target parameters.

[0091] As another implementation, step S130 above includes: selecting multiple sets of target advertising data from multiple sets of target advertising data that have the same combination of markers as the advertising data to be delivered and that have the same at least one dimension information as the advertising data to be delivered, and obtaining target parameter cumulative values ​​based on the parameter cumulative values ​​of the multiple sets of target advertising data respectively.

[0092] In this approach, the cumulative target parameter value can be obtained from the cumulative parameter values ​​of the multiple sets of target ad data in two ways: First, by weighted summing of the cumulative parameter values ​​corresponding to the multiple sets of target ad data. Second, by averaging the cumulative parameter values ​​corresponding to the multiple sets of target ad data.

[0093] Specifically, when using the weighted sum of the cumulative values ​​of parameters corresponding to the multiple sets of target advertising data as the cumulative value of the target parameters, weights can be assigned to the groups of advertising data according to the priority of the dimension objects to which the dimension information of the multiple groups of target advertising data belongs, and the cumulative value of the target parameters can be obtained by weighted summing according to the cumulative values ​​of parameters corresponding to the multiple groups of target advertising data and the weights.

[0094] As another implementation, step S130 above further includes: selecting at least one set of target advertising data from multiple sets of target advertising data that has the same combination of markers as the advertising data to be delivered and is the same as at least one dimension information of the advertising data to be delivered; selecting a set of target advertising data from at least one set of target advertising data according to the priority of the dimension object to which the dimension information of the at least one set of target advertising data belongs; and using the parameter accumulation value corresponding to the set of target advertising data as the target parameter accumulation value.

[0095] Step S140: Obtain the calibration coefficient of the advertising data to be delivered based on the cumulative value of the target parameter.

[0096] When the cumulative value of the target parameter includes at least one of the cumulative value of the target reported conversions, the cumulative value of the target estimated conversion rate, the cumulative value of the target estimated click-through rate, and the cumulative value of the target clicks, there can be multiple ways to obtain the calibration number based on the cumulative value of the target parameter.

[0097] In one implementation, step S140 may involve numerically calculating at least one of the target reported conversion cumulative value, target estimated conversion rate cumulative value, target estimated click-through rate cumulative value, and target click cumulative value, among other things, to obtain a calibration coefficient.

[0098] The numerical calculations mentioned above can be weighted calculations or calculations using pre-set formulas; no specific limitations are made here.

[0099] As another implementation, step S140 above may also involve performing numerical calculations on at least one of the target reported conversion cumulative value, target estimated conversion rate cumulative value, target estimated click-through rate cumulative value, and target click cumulative value, to obtain an initial calibration coefficient; determining the exposure stage and / or consumption stage of the advertising data to be deployed based on the parameter cumulative value in the preset market; determining an expansion coefficient based on the exposure stage and / or consumption stage; and using the expansion coefficient and the initial calibration coefficient to obtain the calibration coefficient of the advertising data to be deployed.

[0100] Step S150: Use the calibration coefficient to calibrate the estimated conversion rate of the advertising data to be delivered, and obtain the calibrated estimated conversion rate.

[0101] There are several ways to calibrate the estimated conversion rate of the advertising data to be delivered using the calibration coefficient.

[0102] One approach is to superimpose the calibration coefficient with the estimated conversion rate of the advertising data to be delivered to obtain the calibrated estimated conversion rate.

[0103] As another implementation method, the calibration coefficient can be multiplied by the estimated conversion rate of the advertising data to be delivered to obtain the calibrated estimated conversion rate.

[0104] Specifically, such as Figure 3 As shown, Figure 3 The diagram illustrates M ad data points that make up 10 ad data points. Each ad data point includes three dimensions: advertiser, brand of the advertised product, and product. The identifier for each ad data point consists of an ad playback platform identifier and an ad launch information identifier.

[0105] If the advertising data to be delivered included in the M advertising data is advertising data 5, then when selecting the target parameter cumulative value from the parameter cumulative values ​​corresponding to multiple sets of target advertising data within a preset time period based on the advertising data to be delivered and the identifier combination corresponding to the advertising data to be delivered, the number of target parameter cumulative values ​​is three, and the set of target advertising data corresponding to each target parameter cumulative value is advertising data with the same dimension object and the same identifier combination as the advertising data to be delivered.

[0106] Accordingly, the three sets of target advertising data corresponding to the cumulative values ​​of the three parameters mentioned above are the first set of target advertising data that has the same dimension information as the advertiser dimension of the advertising data to be placed and has the same identifier combination, the second set of target advertising data that has the same dimension information as the advertising product brand dimension of the advertising data to be placed and has the same identifier combination, and the third set of target advertising data that has the same dimension information as the advertising product dimension of the advertising data to be placed and has the same identifier combination.

[0107] like Figure 4 As shown, Figure 4 The first set of target advertising data includes the advertising data. Specifically, the first set of target advertising data consists of advertising data that shares the same advertiser dimension as the advertising data to be delivered (Advertiser B), is published on Platform A, and is identified as an old advertisement (the first set of target advertising data includes advertising data 5 and advertising data 7).

[0108] like Figure 5 As shown, Figure 5 The first set of target advertising data includes the advertising data. The second set of target advertising data consists of advertising data that shares the same brand information (brand D) as the advertising product brand data to be delivered, and is published on platform A, with the advertising launch information identified as an old advertisement (i.e., the second set of target advertising data includes advertising data 2, advertising data 5, and advertising data 10).

[0109] like Figure 6 As shown, Figure 6The third set of target advertising data includes the advertising data. Specifically, the third set of target advertising data consists of advertising data whose dimensional information for the product dimension is the same as the advertising data to be delivered (product F), and whose publishing platform is platform A, and whose online information is identified as an old advertisement (i.e., the third set of target advertising data includes advertising data 5, advertising data 7, advertising data 9, and advertising data 10).

[0110] When obtaining the calibration coefficient for the advertising data to be delivered based on the cumulative value of the target parameters, and then using the calibration coefficient to calibrate the estimated conversion rate of the advertising data to be delivered, the calibration coefficient can be calculated based on the cumulative value of one or more of the corresponding parameters from the first group of target advertising data, the second group of target advertising data, and the third group of target advertising data. This calibration coefficient is then used to calibrate the estimated conversion rate of the advertising data to be delivered. Figures 4-6 It can be seen that the advertising data included in the first set of target advertising data, the second set of target advertising data, and the third set of target advertising data have a high degree of similarity and correlation with advertising data 5. Accordingly, the cumulative parameter values ​​corresponding to the first set of target advertising data, the second set of target advertising data, and the third set of target advertising data are obtained respectively, and the estimated conversion rate corresponding to advertising data 5 is calibrated using the cumulative parameter values ​​obtained above.

[0111] By employing the advertising conversion rate calibration method provided in this application, when selecting target parameter cumulative values ​​from multiple sets of target advertising data corresponding to parameter cumulative values ​​within a preset time period, the selected target parameter cumulative values ​​are those corresponding to target advertising data that have the same dimensional information and corresponding to the same dimensional combination as the advertising data to be deployed. That is, the target advertising data corresponding to the target parameter cumulative values ​​are highly correlated with or similar to the advertising data to be deployed. Therefore, obtaining calibration coefficients based on the target parameter cumulative values ​​and using these calibration coefficients to calibrate the estimated conversion rate results in a more accurate calibrated estimated conversion rate.

[0112] Please see Figure 7 Another embodiment of this application provides a method for calibrating advertising conversion rates, the method comprising:

[0113] Step S210: Obtain M advertising data and the corresponding identifier combination for each advertising data.

[0114] Among them, the M advertising data include advertising data to be delivered. Each advertising data includes N dimension information, and each dimension information corresponds to a different dimension object. M is an integer greater than 1, and N is an integer greater than 1.

[0115] Step S220: Obtain the cumulative value of the first parameter related to the ad conversion rate of multiple groups of target ad data within the first preset duration, and the cumulative value of the second parameter related to the ad conversion rate of multiple groups of target ad data within the second preset duration.

[0116] Among them, the ad consumption value ranges corresponding to the respective ad data included in each group of target ad data are the same.

[0117] The ad consumption value range can be set in advance, and its setting method can be determined according to the preset ad consumption threshold or according to the conversion bid of the ad, and no specific limitation is made here.

[0118] As an implementation manner, if it is determined according to the conversion bid of the ad, for example, the above ad consumption value range includes: Cost <= 4 * target_cpa, 4 * target_cpa < Cost <= 10 * target_cpa, 10 * target_cpa < Cost <= 20 * target_cpa, and 20 * target_cpa < cost, where Cost refers to the ad consumption and target_cpa refers to the conversion bid of the ad data to be put on the market.

[0119] As another implementation manner, if it is determined according to the preset ad consumption threshold, for example, if the ad consumption threshold is 4 * target_cpa, the ad consumption value range includes Cost <= 4 * target_cpa and 4 * target_cpa < Cost.

[0120] The first preset duration is less than the second preset duration. Each group of target ad data has the same dimension information and corresponds to the same identification combination under at least one dimension object. The target ad data is selected from M ad data.

[0121] Specifically, the first preset duration can be one hour, two hours, twelve hours, or one day, etc. It should be understood that when the first preset duration is one hour, the second preset duration can be twelve hours or one day, etc.

[0122] As an implementable manner, the first preset duration is one hour and the second preset duration is one day.

[0123] Step S230: Select the first initial parameter cumulative value corresponding to each dimension information of the ad data to be put on the market from the cumulative values of the first parameter corresponding to multiple groups of target ad data.

[0124] Among them, the identification combination corresponding to the first initial parameter cumulative value is the identification combination corresponding to the ad data to be put on the market.

[0125] Regarding the specific selection method for choosing the first initial parameter cumulative value corresponding to each dimension of the advertising data to be delivered from the first parameter cumulative values ​​corresponding to multiple sets of target advertising data, please refer to the detailed description of step S130 above, which will not be repeated here.

[0126] Step S240: Detect whether there is a first target consumption accumulation value among the accumulated values ​​of each first initial parameter.

[0127] Among them, the minimum value of the advertising consumption range corresponding to the first target cumulative consumption value is greater than the preset advertising consumption threshold.

[0128] If it does not exist, proceed to step S250: Based on the advertising data to be delivered, the identifier combination corresponding to the advertising data to be delivered, and the priority order of the N dimension objects, select the target parameter cumulative value from the second parameter cumulative values ​​corresponding to the multiple sets of target advertising data.

[0129] It should be understood that, since the second duration is longer than the first duration, the ad spending generated by each ad data in the same set of target ad data within the second preset duration is usually greater than the ad spending within the first preset duration.

[0130] The priority order of the N dimension objects can be preset according to actual needs. For instance, if the N dimension objects include the advertiser dimension, the advertised product brand dimension, and the advertised product dimension, then the corresponding priority order could be advertiser dimension > advertised product dimension > advertised product brand dimension.

[0131] As one implementation method, the method for selecting the target parameter cumulative value in step S250 above can be: selecting the second initial parameter cumulative value from multiple sets of target advertising data and the identifier combination corresponding to the advertising data to be delivered and the advertising data to be delivered; and selecting the target parameter cumulative value from each second target parameter cumulative value according to the dimension information corresponding to each second target parameter cumulative value, the dimension object to which the dimension information belongs, and the priority order of N dimension objects.

[0132] If a first target parameter cumulative value exists among the cumulative values ​​of each first initial parameter, then step S260 is executed: obtain the number of the first target parameter cumulative values.

[0133] Step S270: If the first target parameter has a cumulative value of one, then the first target parameter cumulative value is used as the target parameter cumulative value.

[0134] Step S280: If there are at least two first target parameter cumulative values, then select a target parameter cumulative value from the at least two first target parameter cumulative values ​​according to the dimension information corresponding to the at least two first target parameter cumulative values, the dimension object to which the dimension information belongs, and the priority order of the N dimension objects.

[0135] Please refer to it again. Figures 3-6 and with Figures 3-6 According to the relevant description, if the selected first parameter cumulative value includes the parameter cumulative value corresponding to the first group of target advertising data, the parameter cumulative value corresponding to the second group of target advertising data, and the parameter cumulative value corresponding to the third group of target advertising data, and the priority order of the N dimension objects is advertiser dimension > advertising product dimension > advertising product brand dimension, then the final selected parameter cumulative value (target parameter cumulative value) is the parameter cumulative value corresponding to the first group of target advertising data.

[0136] Step S290: Obtain the calibration coefficient of the advertising data to be delivered based on the cumulative value of the target parameter.

[0137] Step S300: Use the calibration coefficient to calibrate the estimated conversion rate of the advertising data to be delivered, and obtain the calibrated estimated conversion rate.

[0138] Please refer to the following: Figure 8 This application selects P1 group of first target advertising data and P2 group of second target advertising data from M advertising data based on the dimension objects of M advertising data and the corresponding identifier combination for each advertising data. Each group of first target advertising data has the same dimensional information and corresponds to the same identifier combination under at least one dimension object, and the advertising consumption of each advertising data in each group of first target advertising data falls within the same advertising consumption value range within a first preset time period. Similarly, each group of second target advertising data has the same dimensional information and corresponds to the same identifier combination under at least one dimension object, and the advertising consumption of each advertising data in each group of second target advertising data falls within the same advertising consumption value range within a second preset time period.

[0139] For the first target ad data in group P1, determine whether there is at least one set of first designated target ad data in the ad consumption range corresponding to the first target ad data in group P1 where the minimum value of the consumption range is greater than the preset ad consumption threshold. If there is a first designated target ad data in group Q1, select a set of first target ad data from the first designated target ad data in group Q1 according to the priority order of the dimension objects and N dimension objects corresponding to the first designated target ad data in group Q1, and use the cumulative value of the parameters corresponding to the first target ad data as the cumulative value of the target parameters.

[0140] If not, for the second target ad data of group P2, determine whether there is at least one set of second designated target ad data in the ad consumption range corresponding to the second target ad data of group P2 where the minimum value of the consumption range is greater than the preset ad consumption threshold. If there is a second designated target ad data of group Q2, select a set of second target ad data from the second designated target ad data of group Q2 according to the priority order of the dimension object and N dimension objects corresponding to the second designated target ad data of group Q2, and use the parameter accumulation value corresponding to the second target ad data as the target parameter accumulation value.

[0141] By adopting the above selection method, the range of target advertising data corresponding to the cumulative value of the target parameter can be gradually narrowed down so that the final selected cumulative value of the target parameter is the set of target advertising data with the strongest correlation or the highest similarity to the advertising data to be delivered.

[0142] The method for calibrating advertising conversion rates provided in this application, by adopting the above steps S230-S280, can effectively ensure that the data of each advertising data in a set of target advertising data corresponding to the selected target parameter cumulative value is valid and reliable, and is advertising data that is highly correlated with or similar to the advertising data to be placed. Accordingly, the calibration coefficient is obtained based on the target parameter cumulative value, and the calibrated estimated conversion rate obtained by calibrating the estimated conversion rate using the calibration coefficient is more accurate.

[0143] Please see Figure 9 This application provides a method for calibrating advertising conversion rates, including:

[0144] Step S310: Obtain M ad data points and the corresponding identifier combination for each ad data point.

[0145] Among them, the M advertising data include advertising data to be delivered. Each advertising data includes N dimension information, and each dimension information corresponds to a different dimension object. M is an integer greater than 1, and N is an integer greater than 1.

[0146] Step S320: Obtain the cumulative values ​​of parameters related to the ad conversion rate for multiple sets of target ad data within a preset time period.

[0147] Each set of target ad data has the same dimensional information and the same identifier combination under at least one dimensional object. The target ad data is selected from M ad data.

[0148] Step S330: Based on the advertising data to be delivered and the corresponding identifier combination, select the target parameter cumulative value from the parameter cumulative values ​​corresponding to multiple sets of target advertising data within a preset time period.

[0149] The cumulative values ​​of the target parameters include the cumulative value of the target reported conversions, the cumulative value of the target estimated conversion rate, the cumulative value of the target estimated click-through rate, and the cumulative value of the target clicks.

[0150] Step S340: Perform numerical calculations on the cumulative value of the target reported conversions, the cumulative value of the target estimated conversion rate, the cumulative value of the target estimated click-through rate, and the cumulative value of the target clicks to obtain the initial calibration coefficients for the advertising data to be deployed.

[0151] The methods for calculating the cumulative values ​​of reported conversions, estimated conversion rates, estimated click-through rates, and clicks can be either fixed calculation methods such as addition or multiplication, or simply adding these values ​​together. Alternatively, they can be calculated using pre-set formulas. No specific limitations are imposed here; the calculations can be tailored to actual needs.

[0152] As one possible implementation method, the initial calibration coefficient is calculated by the following formula for the cumulative value of the target reported conversions, the cumulative value of the target estimated conversion rate, the cumulative value of the target estimated click-through rate, and the cumulative value of the target clicks: cali_rate=Conv_valid / (PCVR_valid*PCTR_valid / ClickNum_valid), where cali_rate is the initial calibration coefficient, Conv_valid is the cumulative value of the target reported conversions, PCVR_valid is the cumulative value of the target estimated conversion rate, PCTR_valid is the cumulative value of the target estimated click-through rate, and ClickNum_valid is the cumulative value of the target clicks.

[0153] Step S350: Determine the exposure stage of the advertising data to be delivered based on the cumulative consumption value and the cumulative reported conversion value of the advertising data within the preset time period.

[0154] The server can store multiple exposure stages and the corresponding cumulative consumption value range and cumulative reported conversion value range for each exposure stage.

[0155] As one implementation method, the multiple exposure stages include a first exposure stage and a second exposure stage. If the cumulative consumption value of the advertising data to be delivered within a preset time period is less than a preset cumulative consumption threshold, and the cumulative reported conversion value of the advertising data to be delivered within a preset time period is less than a preset conversion threshold, then the exposure stage of the advertising data to be delivered is determined to be the first exposure stage.

[0156] If the cumulative consumption value of the advertising data to be delivered within the preset time period is not less than the preset cumulative consumption threshold, or the cumulative reported conversion value of the advertising data to be delivered within the preset time period is not less than the preset conversion threshold, then the exposure stage of the advertising data to be delivered is determined as the second exposure stage.

[0157] It should be understood that multiple exposure stages may also include a third or fourth exposure stage, without specific limitations here.

[0158] In this implementation, the specific values ​​of the preset consumption accumulation value and the preset conversion accumulation value can be set according to actual needs.

[0159] For example, the preset cumulative spending value is 2 * target_cpa, and the preset ad conversion count is 2. That is, if the cumulative spending value of the ad data to be delivered within the preset time period is less than 2 * target_cpa, and the cumulative reported conversion count of the ad data to be delivered within the preset time period is less than 2, the exposure stage of the ad data to be delivered is determined to be the first exposure stage (initial exposure stage). If the cumulative spending value of the ad data to be delivered within the preset time period is greater than or equal to 2 * target_cpa, or the cumulative reported conversion count of the ad data to be delivered within the preset time period is greater than or equal to 2, then the exposure stage of the ad data to be delivered is determined to be the second exposure stage (exposure maturity stage).

[0160] Step S360: Determine the consumption stage of the advertising data to be delivered based on the cumulative consumption value of the advertising data to be delivered.

[0161] The server can store multiple consumption stages and the cumulative consumption value range corresponding to each exposure stage.

[0162] In one implementation, the consumption stages of the advertising data to be delivered may include a first consumption stage, a second consumption stage, and a third consumption stage. Specifically, the cumulative consumption value for the first consumption stage ranges from [0, 8*target_cpa], the cumulative consumption value for the second consumption stage ranges from [8*target_cpa, 25*target_cpa], and the cumulative consumption value for the third consumption stage ranges from greater than 25*target_cpa.

[0163] Step S370: Obtain the magnification factor based on the exposure stage and the consumption stage.

[0164] It should be understood that different exposure and consumption stages can correspond to different scaling factors. That is, the server can store multiple scaling factors and the corresponding consumption and exposure stages for each scaling factor. Different exposure and consumption stages can also correspond to different scaling factor calculation methods; that is, the server can also store different scaling factor calculation methods for different exposure and consumption stages. The settings can be configured according to actual needs.

[0165] As one possible implementation, when the exposure stage is the first exposure stage and the consumption stage is the first consumption stage, a preset constant is obtained, which is the expansion coefficient of the initial calibration coefficient.

[0166] The aforementioned preset length constant can be any one of 0.9, 0.95, 1, and 1.05.

[0167] Considering that the consumption in the first consumption stage is relatively low, such as if the cumulative consumption of the advertising data to be delivered within the preset time is less than 8*target_cpa, the preset constant is 1, that is, no expansion is performed.

[0168] When the exposure phase is the first exposure phase and the consumption phase is the second consumption phase, the deviation value of the advertising data to be delivered is obtained based on the cumulative consumption value and cumulative conversion value of the advertising data to be delivered within a preset time period and the conversion bid of the advertising data to be delivered. The first deviation correction parameter is obtained based on the cumulative consumption value of the advertising data to be delivered and the first target consumption value in the consumption range corresponding to the second consumption phase. The expansion coefficient is obtained based on the first deviation correction parameter and the deviation value. The maximum consumption value in the consumption range corresponding to the first consumption phase is less than the minimum consumption value in the consumption range corresponding to the second consumption phase.

[0169] For example, the cumulative consumption value range for the second consumption stage is [8*target_cpa, 25*target_cpa]. The scaling factor can be calculated using the following formula:

[0170]

[0171] Where cost is the cumulative cost of the ads to be delivered within the preset time period, and current_cpa_bias is the current CPA bias of the ads. comv_mum refers to the cumulative number of conversions of the advertisement to be delivered within a preset time period.

[0172] When the exposure phase is the first exposure phase and the consumption phase is the third consumption phase, the deviation value of the advertising data to be delivered is obtained based on the cumulative consumption value and cumulative conversion value of the advertising data to be delivered within the preset time period and the conversion bid of the advertising data to be delivered. The second deviation correction parameter is obtained based on the second target consumption value and the third consumption value in the consumption range corresponding to the third consumption phase. The expansion coefficient is obtained based on the second deviation correction parameter and the deviation value. The maximum consumption value in the consumption value range corresponding to the second consumption phase is less than the minimum consumption value in the consumption value range corresponding to the third consumption phase.

[0173] For example, the cumulative consumption value corresponding to the third consumption stage is greater than 25 * target_cpa. In this case, the scaling factor can be calculated using the following formula:

[0174]

[0175] Where cost is the cumulative cost of the ad data to be delivered within the preset time period, and current_cpa_bias is the current CPA bias of the ad. comv_mum refers to the cumulative number of conversions of the advertising data to be delivered within a preset time period.

[0176] When the exposure stage is the second exposure stage, step S370 above also includes:

[0177] Step S372: Obtain the smoothing coefficient of the advertising data to be deployed based on the historical cumulative conversion value and preset duration.

[0178] The historical cumulative conversion value of the advertising data to be placed can be the total number of conversions that occurred in the day, two days or one week before the current time, or the total number of conversions that occurred in the day, two days or one week before the current time. There is no specific limitation here.

[0179] The preset duration mentioned above can be in units of one day or in units of hours.

[0180] For example, the market is assumed to be measured in hours, and the smoothing coefficient of the advertising data to be delivered can be obtained using the following formula:

[0181]

[0182] Where, smooth_base is the smoothing coefficient, ceil is the rounding function, sum_conversion_yesterday refers to the historical conversion data of the ad to be delivered, and H refers to the value corresponding to the preset duration in hours.

[0183] Step S374: Based on the historical conversion volume and historical conversion rate of the advertising data to be deployed, the estimated cumulative value of the historical conversion volume and historical conversion rate, the smoothing coefficient, and the cumulative value of the target parameter, obtain the expansion coefficient of the advertising data to be deployed.

[0184] For example, the scaling factor of the advertising data to be delivered can be calculated using the following formula:

[0185]

[0186] Wherein, calibration_rate is the amplification factor of the ad data to be delivered in the second exposure stage, hisrestore_pcvr_bias_factord=(Conv_valid_p / PCVR__valid_p) / (Conv_valid / PCVR__valid), where Conv_valid_p is the cumulative target conversion value corresponding to the consumption stage of the ad data to be delivered, PCVR__valid_p is the estimated cumulative conversion rate value corresponding to the consumption stage of the ad data to be delivered, Conv_valid is the cumulative target conversion value included in the cumulative target parameter value, PCVR__valid is the cumulative target conversion rate value included in the cumulative target parameter value, coversion_num is the historical conversion volume of the ad data to be delivered, sum_pcvr is the cumulative conversion rate value of the ad data to be delivered within the preset duration, sum_pctr is the estimated cumulative click-through rate value of the ad data to be delivered within the preset duration, and clicknum is the number of clicks of the ad data to be delivered within the preset duration.

[0187] It should be noted that Conv_valid_p is obtained by selecting ad data from the target ad data set corresponding to the cumulative value of the target parameter, which is in the same consumption stage as the ad data to be delivered, and obtaining the cumulative conversion value of the ad data combination formed by this ad data within a preset time period. Similarly, PCVR_valid_p is obtained by selecting ad data from the target ad data set corresponding to the cumulative value of the target parameter, which is in the same consumption stage as the ad data to be delivered, and obtaining the estimated cumulative conversion rate value of the ad data combination formed by this ad data within a preset time period.

[0188] As can be seen from the formula for calculating the smoothing coefficient, the larger the cumulative historical conversion value, the larger the value of smooth_base. Correspondingly, the shorter the continuous exposure time (preset duration) of the ad data to be delivered on the day, the smaller the value of smooth_base. By introducing a smoothing coefficient, data anomalies caused by a small amount of historical exposure data corresponding to the ad data to be delivered can be effectively avoided, which could lead to abnormally high or low calculated calibration coefficients.

[0189] Step S380: Process the initial calibration coefficients according to the expansion coefficients to obtain the calibration coefficients of the advertising data to be delivered.

[0190] The method of processing the initial calibration coefficient based on the expansion coefficient to obtain the calibration coefficient of the advertising data to be placed can be either by adding the expansion coefficient to the initial calibration coefficient to obtain the calibration coefficient of the advertising data to be placed, or by multiplying the expansion coefficient to the initial calibration coefficient to obtain the calibration coefficient of the advertising data to be placed.

[0191] As one implementation, step S370 includes multiplying the expansion coefficient by the initial calibration coefficient to obtain the calibration coefficient of the advertising data to be delivered.

[0192] Step S390: Use the calibration coefficient to calibrate the estimated conversion rate of the advertising data to be delivered, and obtain the calibrated estimated conversion rate.

[0193] The present application provides a method for calibrating advertising conversion rates. By adopting the above steps S330-S380, after selecting the cumulative value of the target parameter, the calibration coefficients of the advertising data to be delivered under different exposure stages and different consumption stages can be accurately obtained based on the cumulative value of the target parameter, the exposure stage and the consumption stage of the advertising data to be delivered. Therefore, when using the above-obtained calibration coefficients to calibrate the estimated conversion rate of the advertising data to be delivered, the calibrated estimated conversion rate is more accurate.

[0194] Please see Figure 10 This application also provides a method for calibrating advertising conversion rates, the method comprising:

[0195] Step S410: Obtain M advertising data and the corresponding identifier combination for each advertising data.

[0196] Among them, the M advertising data include advertising data to be delivered. Each advertising data includes N dimension information, and each dimension information corresponds to a different dimension object. M is an integer greater than 1, and N is an integer greater than 1.

[0197] Step S420: Group the M advertising data according to the combination of dimension objects and identifiers to obtain multiple groups of target advertising data, and obtain the cumulative value of the first parameter related to the advertising conversion rate of the multiple groups of target advertising data within a first preset time period, and the cumulative value of the second parameter related to the advertising conversion rate of the multiple groups of target advertising data within a second preset time period.

[0198] Each set of target ad data has the same dimensional information and the same identifier combination under at least one dimensional object. The target ad data is selected from M ad data.

[0199] When grouping M advertising data points according to the combination of dimension objects and identifiers, the specific distinction is made between the advertiser dimension, the advertising product dimension, and the advertising product brand dimension.

[0200] At the advertiser level, the system calculates the cumulative parameter values ​​(cumulative reported conversions, cumulative estimated conversion rate, cumulative estimated click-through rate, and cumulative clicks) for different advertisers under different identifier dimensions (e.g., [Platform I, New Ad], [Platform I, Old Ad], [Platform J, New Ad], and [Platform J, Old Ad]) within a preset time period. Taking a preset time period of one day as an example, the cumulative parameter values ​​within a day can be obtained by accumulating them using a time decay strategy.

[0201] Specifically, the cumulative number of conversions reported by different advertisers within a day is obtained using the following formula: In the formula, Conv_advertiser_day is the cumulative number of reported conversions for the same advertiser within one day under the advertiser dimension, I is the current hour, lambada is the time decay coefficient, which is a constant value (e.g., 0.05), and conv_advertiser_hour i This represents the total number of reported conversions in the i-th hour.

[0202] The estimated cumulative conversion rate for different advertisers within a day is obtained using the following formula: In the formula, PCVR_advertiser_day is the estimated cumulative conversion rate of the same advertiser within one day under the advertiser dimension, I is the current hour, lambada is the time decay coefficient, which is a constant value (e.g., 0.05), and PCVR_advertiser_hour is the time decay coefficient. i The total estimated conversion rate is calculated at time i.

[0203] The cumulative number of clicks for different advertisers within a day is obtained using the following formula: In the formula, ClickNum_advertiser_day is the cumulative number of clicks for the same advertiser within one day, I is the current hour, lambda is the time decay coefficient, which is a constant value (e.g., 0.05), and ClickNum_advertiser_hour is the number of clicks. i This represents the total number of clicks counted at time i.

[0204] The estimated cumulative click-through rate (CTR) for different advertisers within a day is obtained using the following formula: In the formula, PCTR_advertiser_day is the estimated cumulative click-through rate of the same advertiser within one day under the advertiser dimension, I is the current hour, lambada is the time decay coefficient, which is a constant value (e.g., 0.05), and PCTR_advertiser_hour is the time decay coefficient. i This represents the total cumulative estimated click-through rate obtained at time i.

[0205] By employing a similar approach, it is also possible to statistically analyze the cumulative parameter values ​​(cumulative reported conversion count, cumulative estimated conversion rate, cumulative estimated click-through rate, and cumulative click count) for different advertising product brands under different identifier dimensions (e.g., [Platform I, New Ad], [Platform I, Old Ad], [Platform J, New Ad], and [Platform J, Old Ad]) within a preset time period. Furthermore, it is possible to statistically analyze the cumulative parameter values ​​(cumulative reported conversion count, cumulative estimated conversion rate, cumulative estimated click-through rate, and cumulative click count) for different products under different identifier dimensions (e.g., [Platform I, New Ad], [Platform I, Old Ad], [Platform J, New Ad], and [Platform J, Old Ad]) within a preset time period.

[0206] It should be understood that if further differentiation of spending tiers is needed, the cumulative parameter values ​​for each dimension object under different spending dimensions, under different dimension information and different label dimensions, can be further calculated within a preset time period. For example, when the dimension object is the advertiser dimension, the cumulative parameter values ​​for different advertisers under different label dimensions can be further calculated within a preset time period.

[0207] By adopting the above method, multiple sets of target advertising data can be used to accumulate the first parameter value within a first preset duration (one hour) and the second parameter value within a second preset duration (one day). Each set of target advertising data has the same dimensional information and corresponds to the same identifier combination under at least one dimensional object.

[0208] Step S430: Determine whether the cumulative value of the first parameter related to the ad conversion rate is sufficient within the first preset time period.

[0209] Specifically, the method for determining whether the first parameter cumulative value is sufficient can be as follows: From the first parameter cumulative values ​​corresponding to multiple sets of target advertising data, select the first initial parameter cumulative value corresponding to each dimension of the advertising data to be delivered. The identifier combination corresponding to the first initial parameter cumulative value is the identifier combination corresponding to the advertising data to be delivered. Check whether a first target consumption cumulative value exists among the first initial parameter cumulative values, wherein the minimum value of the advertising consumption range corresponding to the first target consumption cumulative value is greater than a preset advertising consumption threshold. If a first target consumption cumulative value exists, the first parameter cumulative value within the first preset time period is sufficient; if no first target consumption cumulative value exists, it is insufficient.

[0210] If sufficient, proceed to step S440 to determine the target parameter cumulative value based on the first parameter cumulative value.

[0211] Specifically, if there is one first target parameter cumulative value, then the first target parameter cumulative value is used as the target parameter cumulative value; if there are at least two first target parameter cumulative values, then the target parameter cumulative value is selected from the at least two first target parameter cumulative values ​​according to the dimension information corresponding to the at least two first target parameter cumulative values, the dimension object to which the dimension information belongs, and the priority order of the N dimension objects.

[0212] If insufficient, proceed to step S450: Based on the advertising data to be delivered, the identifier combination corresponding to the advertising data to be delivered, and the priority order of the N dimension objects, select the target parameter cumulative value from the cumulative values ​​of the second parameter corresponding to the multiple sets of target advertising data.

[0213] Among them, the priority order of the N dimensions is advertiser dimension > advertised product dimension > advertised product brand dimension.

[0214] Step S460: Calculate the calibration coefficient of the advertising data to be delivered based on the cumulative value of the target parameters.

[0215] Specifically, during calibration, the initial calibration coefficients for the target reported cumulative conversions, target estimated conversion rate, target estimated click-through rate, and target cumulative clicks can be calculated using the following formula: cali_rate=Conv_valid / (PCVR_valid*PCTR_valid / ClickNum_valid), where cali_rate is the initial calibration coefficient, Conv_valid is the target reported cumulative conversions, PCVR_valid is the target estimated conversion rate, PCTR_valid is the target estimated click-through rate, and ClickNum_valid is the target cumulative clicks.

[0216] After obtaining the initial calibration coefficients, the exposure and consumption stages of the ad data to be delivered can be determined based on the ad spend and reported conversions within a preset time period. Specifically, if the cumulative spend of the ad data to be delivered within the preset time period is less than 2 * target_cpa, and the cumulative reported conversions within the preset time period are less than 2, the exposure stage of the ad data to be delivered is determined to be the first exposure stage (initial exposure stage). If the cumulative spend of the ad data to be delivered within the preset time period is greater than or equal to 2 * target_cpa, or the cumulative reported conversions within the preset time period are greater than or equal to 2, the exposure stage of the ad data to be delivered is determined to be the second exposure stage (exposure maturity stage). The consumption stages of the ad delivery can include the first consumption stage, the second consumption stage, and the third consumption stage. The cumulative spend range for the first consumption stage is [0, 8 * target_cpa], the cumulative spend range for the second consumption stage is [8 * target_cpa, 25 * target_cpa], and the cumulative spend range for the third consumption stage is greater than 25 * target_cpa.

[0217] After obtaining the exposure and consumption phases of the ads to be placed, an amplification factor can be calculated based on these phases. Specifically:

[0218] If the exposure stage of the ad data to be delivered is the first exposure stage (initial exposure stage), and the consumption stage is the first consumption stage, considering that the consumption in the first consumption stage is relatively low, if the cumulative consumption value of the ad data to be delivered within the preset time is less than 8*target_cpa, the preset constant is 1, that is, no expansion is performed.

[0219] When the exposure stage of the ad data to be delivered is the first exposure stage (initial exposure stage), and the consumption stage is the second stage, the cumulative consumption value range corresponding to the second consumption stage is [8*target_cpa, 25*target_cpa]. At this time, the amplification factor can be calculated using the following formula:

[0220]

[0221] Where cost is the cumulative cost of the ad data to be delivered within the preset time period, and current_cpa_bias is the current CPA bias of the ad. comv_mum refers to the cumulative number of conversions of the advertisement to be delivered within a preset time period.

[0222] If the exposure stage of the ad data to be delivered is the first exposure stage (initial exposure stage), and the consumption stage is the third stage, then the cumulative consumption value corresponding to the third consumption stage is greater than 25 * target_cpa. In this case, the scaling factor can be calculated using the following formula:

[0223]

[0224] Where cost is the cumulative cost of the ad data to be delivered within the preset time period, and current_cpa_bias is the current CPA bias of the ad. comv_mum refers to the cumulative number of conversions of the advertising data to be delivered within a preset time period.

[0225] By using the above method, the amplification factor corresponding to different consumption stages can be obtained under the initial exposure stage.

[0226] When the data for the ads to be delivered has moved beyond the initial exposure stage (first exposure stage) and entered the mature stage (second exposure stage) where exposure is relatively sufficient, different calibration strategies need to be applied. At this point, a smoothing coefficient can be calculated based on the historical cumulative conversion value and preset duration of the ad data, using the following formula, where the formula includes:

[0227]

[0228] Where, smooth_base is the smoothing coefficient, ceil is the rounding function, sum_conversion_yesterday refers to the historical conversion data of the ad to be delivered, and H refers to the value corresponding to the preset duration in hours.

[0229] After obtaining the smoothing coefficient, the scaling factor of the advertising data to be delivered can be calculated using the following formula:

[0230] Wherein, calibration_rate is the amplification factor of the ad data to be delivered in the second exposure stage, hisrestore_pcvr_bias_factord=(Conv_valid_p / PCVR__valid_p) / (Conv_valid / PCVR__valid), where Conv_valid_p is the cumulative target conversion value corresponding to the consumption stage of the ad data to be delivered, PCVR__valid_p is the estimated cumulative conversion rate value corresponding to the consumption stage of the ad data to be delivered, Conv_valid is the cumulative target conversion value included in the cumulative target parameter value, PCVR__valid is the cumulative target conversion rate value included in the cumulative target parameter value, coversion_num is the historical conversion volume of the ad data to be delivered, sum_pcvr is the cumulative conversion rate value of the ad data to be delivered within the preset duration, sum_pctr is the estimated cumulative click-through rate value of the ad data to be delivered within the preset duration, and clicknum is the number of clicks of the ad data to be delivered within the preset duration.

[0231] Step S470: Multiply the calibration coefficient by the estimated conversion rate of the advertising data to be delivered to obtain the calibrated estimated conversion rate.

[0232] Please see Figure 11 , Figure 11 The data shows the difference between the estimated and actual conversion rates of the advertisements to be delivered under the advertising product dimension, obtained at different time periods, before and after calibration. V31 represents the difference between the estimated and actual conversion rates before calibration, and V41 represents the difference between the estimated and actual conversion rates after calibration. It can be seen that the difference between the estimated and actual conversion rates after calibration is smaller than the difference between the estimated and actual conversion rates before calibration. Therefore, it can be seen that the estimated conversion rate after calibration is more accurate than the estimated conversion rate before calibration.

[0233] Please see Figure 12 , Figure 12 This shows the distribution range of the ratio of estimated conversion rate to actual conversion rate before and after calibration, under the dimension of advertised products. V32 represents the ratio of estimated conversion rate to actual conversion rate before calibration, and V42 represents the percentage of the ratio of estimated conversion rate to actual conversion rate after calibration across different value ranges. Through analysis... Figure 12 It can be seen that the proportion of calibrated estimated conversion rates with a ratio between 0.8 and 1.2 is significantly higher than the proportion of calibrated estimated conversion rates with a ratio between 0.8 and 1.2 before calibration. Therefore, it can be concluded that the calibrated estimated conversion rate is more accurate than the uncalibrated estimated conversion rate.

[0234] Please see Figure 13 , Figure 13 This data represents the difference between the estimated and actual conversion rates of the ads to be delivered, obtained from different time periods under the advertiser's perspective, before and after calibration. Specifically, V33 represents the difference between the estimated and actual conversion rates before calibration, and V43 represents the difference after calibration. It can be seen that the difference between the estimated and actual conversion rates after calibration is smaller than the difference before calibration. Therefore, the estimated conversion rate after calibration is more accurate than the estimated conversion rate before calibration.

[0235] Please see Figure 14 , Figure 14 The chart shows the distribution range of the ratio of estimated conversion rate to actual conversion rate before and after calibration, categorized by advertiser dimension. V34 represents the ratio of estimated to actual conversion rate before calibration, while V44 represents the proportion of the ratio of estimated to actual conversion rate after calibration across different value ranges. It can be seen that the proportion of the ratio of estimated to actual conversion rate after calibration within the range of 0.8-1.2 is significantly higher than the proportion of the ratio within the range of 0.8-1.2 before calibration. Therefore, it can be concluded that the estimated conversion rate after calibration is more accurate than the estimated conversion rate before calibration.

[0236] Please see Figure 15 This application provides an advertising conversion rate calibration device 500, including: a first acquisition module 510, a second acquisition module 520, a parameter cumulative value selection module 530, a calibration coefficient acquisition module 540, and a conversion rate calibration module 550.

[0237] The first acquisition module 510 is used to acquire M advertising data and the corresponding identifier combination for each advertising data.

[0238] Among them, the M advertising data include advertising data to be delivered. Each advertising data includes N dimension information, and each dimension information corresponds to a different dimension object. M is an integer greater than 1, and N is an integer greater than 1.

[0239] As one implementation method, the dimension objects include at least one of the advertiser dimension, product dimension, and advertiser dimension.

[0240] As one implementation method, the identifier combination corresponding to the advertising data consists of the advertising publishing platform identifier and the advertising launch information identifier in the advertising data.

[0241] The second acquisition module 520 is used to acquire the cumulative values ​​of parameters related to the advertising conversion rate of multiple sets of target advertising data within a preset time period.

[0242] Each set of target ad data has the same dimensional information and the same identifier combination under at least one dimensional object. The target ad data is selected from M ad data.

[0243] The parameter accumulation value selection module 530 is used to select the target parameter accumulation value from the parameter accumulation values ​​corresponding to multiple sets of target advertising data within a preset time period, based on the advertising data to be delivered and the identifier combination corresponding to the advertising data to be delivered.

[0244] As an optional implementation, the parameter accumulation value selection module 530 is specifically used to select the target parameter accumulation value from the parameter accumulation values ​​corresponding to multiple sets of target advertising data within a preset time period, based on the advertising data to be delivered, the identifier combination corresponding to the advertising data to be delivered, and the priority order of N dimension objects.

[0245] In this implementation, the advertising consumption value range corresponding to each advertising data included in each set of target advertising data is the same. The preset duration includes a first preset duration and a second preset duration, and the first preset duration is less than the second preset duration. The second acquisition module 520 is also used to acquire the cumulative value of the first parameter related to the advertising conversion rate of multiple sets of target advertising data within the first preset duration, and the cumulative value of the second parameter related to the advertising conversion rate of multiple sets of target advertising data within the second preset duration.

[0246] Please see Figure 16 The parameter cumulative value selection module 530 includes: a first selection unit 531, a detection unit 532, and a second selection unit 533.

[0247] The first selection unit 531 is used to select, from the first parameter cumulative values ​​corresponding to each dimension of the advertising data to be delivered, the first initial parameter cumulative value corresponding to each of the multiple sets of target advertising data, and the identifier combination corresponding to the first initial parameter cumulative value is the identifier combination corresponding to the advertising data to be delivered.

[0248] The detection unit 532 is used to detect whether there is a first target consumption accumulation value among the accumulated values ​​of each first initial parameter, wherein the minimum value of the advertising consumption range corresponding to the first target consumption accumulation value is greater than the preset advertising consumption threshold.

[0249] The second selection unit 533 is used to select the target parameter cumulative value from the second parameter cumulative values ​​corresponding to multiple sets of target advertising data when there is no first target consumption cumulative value, based on the advertising data to be delivered, the identifier combination corresponding to the advertising data to be delivered, and the priority order of N dimension objects.

[0250] As an optional implementation, the parameter accumulation value selection module 530 further includes a quantity acquisition unit 534 and a parameter determination unit 535.

[0251] The quantity acquisition unit 534 is used to acquire the quantity of the first target parameter accumulation value when there is a first target parameter accumulation value among the first initial parameter accumulation values.

[0252] The parameter determination unit 535 is used to take the first target parameter cumulative value as the target parameter cumulative value when there is only one first target parameter cumulative value; or to select the target parameter cumulative value from the at least two first target parameter cumulative values ​​according to the dimension information corresponding to the at least two first target parameter cumulative values, the dimension object to which the dimension information belongs, and the priority order of the N dimension objects when there are at least two first target parameter cumulative values.

[0253] The calibration coefficient acquisition module 540 is used to obtain the calibration coefficient of the advertising data to be delivered based on the cumulative value of the target parameter.

[0254] Please see Figure 17 In one optional implementation, the cumulative value of the target parameter includes the cumulative value of the target reported conversion number, the cumulative value of the target estimated conversion rate, the cumulative value of the target estimated click-through rate, and the cumulative value of the target click number. The calibration coefficient acquisition module 540 includes: a first calibration coefficient acquisition unit 541, an exposure stage determination unit 542, a consumption stage determination unit 543, an expansion coefficient acquisition unit 544, and a second calibration coefficient acquisition unit 545.

[0255] The first calibration coefficient acquisition unit 541 is used to perform numerical calculations on the cumulative value of the target reported conversion number, the cumulative value of the target estimated conversion rate, the cumulative value of the target estimated click rate, and the cumulative value of the target click number to obtain the initial calibration coefficient of the advertising data to be deployed.

[0256] The exposure stage determination unit 542 is used to determine the exposure stage of the advertising data to be delivered based on the cumulative consumption value and the cumulative reported conversion value of the advertising data to be delivered within a preset time period.

[0257] Consumption stage determination unit 543 is used to determine the consumption stage of the advertising data to be delivered based on the cumulative consumption value of the advertising data to be delivered.

[0258] The magnification factor acquisition unit 544 is used to acquire the magnification factor based on the exposure stage and the consumption stage.

[0259] The second calibration coefficient acquisition unit 545 is used to process the initial calibration coefficient according to the expansion coefficient to obtain the calibration coefficient of the advertising data to be delivered.

[0260] In this implementation, the exposure stage determination unit 542 is further configured to: determine the exposure stage of the advertising data to be delivered as a first exposure stage when the cumulative consumption value of the advertising data to be delivered within a preset time period is less than a preset cumulative consumption threshold, and the cumulative reported conversion value of the advertising data to be delivered within a preset time period is less than a preset conversion threshold; or determine the exposure stage of the advertising data to be delivered as a second exposure stage when the cumulative consumption value of the advertising data to be delivered within a preset time period is not less than a preset cumulative consumption threshold, or the cumulative reported conversion value of the advertising data to be delivered within a preset time period is not less than a preset conversion threshold.

[0261] As one implementation, the amplification coefficient acquisition unit 544 is specifically used to: acquire a preset constant when the exposure stage is the first exposure stage and the consumption stage is the first consumption stage, the preset constant being the amplification coefficient of the initial calibration coefficient; or when the exposure stage is the first exposure stage and the consumption stage is the second consumption stage, obtain a deviation value of the advertising data to be delivered based on the cumulative consumption value and cumulative conversion value of the advertising data to be delivered within a preset time period and the conversion bid of the advertising data to be delivered, obtain a first deviation correction parameter based on the cumulative consumption value of the advertising data to be delivered and the first target consumption value in the consumption range corresponding to the second consumption stage, and obtain an amplification coefficient based on the first deviation correction parameter and the deviation value, wherein the first... The maximum consumption value in the consumption range corresponding to the consumption stage is less than the minimum consumption value in the consumption range corresponding to the second consumption stage; or, when the exposure stage is the first exposure stage and the consumption stage is the third consumption stage, the deviation value of the advertising data to be delivered is obtained based on the cumulative consumption value and cumulative conversion value of the advertising data to be delivered within a preset time period and the conversion bid of the advertising data to be delivered. The second deviation correction parameter is obtained based on the second target consumption value and the third consumption value in the consumption range corresponding to the third consumption stage, and the expansion coefficient is obtained based on the second deviation correction parameter and the deviation value. In this case, the maximum consumption value in the consumption value range corresponding to the second consumption stage is less than the minimum consumption value in the consumption value range corresponding to the third consumption stage.

[0262] As one implementation, the expansion coefficient acquisition unit 544 is also used to, when the exposure stage is the second exposure stage, obtain the smoothing coefficient of the advertising data to be placed based on the historical cumulative conversion value and preset duration of the advertising data to be placed; and obtain the expansion coefficient of the advertising data to be placed based on the historical conversion volume and historical conversion rate estimated cumulative value, smoothing coefficient and target parameter cumulative value of the advertising data to be placed.

[0263] The conversion rate calibration module 550 is used to calibrate the estimated conversion rate of the advertising data to be delivered using calibration coefficients, and obtain the calibrated estimated conversion rate.

[0264] As one implementation, the conversion rate calibration module 550 is also used to multiply the calibration coefficient by the estimated conversion rate of the advertising data to be delivered, so as to obtain the calibrated estimated conversion rate.

[0265] In one implementation, if the second selection unit acquires multiple cumulative values ​​of target parameters, the advertising conversion rate calibration device 500 further includes a weighted calculation module. The weighted calculation module is used to perform a weighted summation of the multiple cumulative values ​​of target parameters to obtain a weighted summation of the cumulative values ​​of target parameters.

[0266] It should be noted that the device embodiments in this application correspond to the aforementioned method embodiments. The specific principles in the device embodiments can be found in the content of the aforementioned method embodiments, and will not be repeated here.

[0267] The following description will be based on 18 pairs of electronic devices provided in this application.

[0268] Please see Figure 18 Based on the advertising conversion rate calibration method provided in the above embodiments, this application embodiment also provides another method including an electronic device 100 capable of performing the aforementioned method. The electronic device 100 can be a server or a terminal device, and the terminal device can be a smartphone, tablet computer, computer, or portable computer, etc. As one approach, the electronic device 100 can be as follows: Figure 1 The server 11 or terminal device 12 shown.

[0269] The electronic device 100 includes a processor 102 and a memory 104. The memory 104 stores programs that can execute the contents of the foregoing embodiments, and the processor 102 can execute the programs stored in the memory 104.

[0270] The processor 102 may include one or more cores for data processing and message matrix units. The processor 102 connects to various parts within the electronic device 100 using various interfaces and lines, and performs various functions and processes data by running or executing instructions, programs, code sets, or instruction sets stored in the memory 104, and by calling data stored in the memory 104. Optionally, the processor 102 may be implemented using at least one hardware form of Digital Signal Processing (DSP), Field-Programmable Gate Array (FPGA), or Programmable Logic Array (PLA). The processor 102 may integrate one or more of the following: Central Processing Unit (CPU), Graphics Processing Unit (GPU), and modem. The CPU primarily handles the operating system, user interface, and applications; the GPU is responsible for rendering and drawing the displayed content; and the modem handles wireless communication. It is understood that the modem may also not be integrated into the processor 102 and may be implemented separately using a communication chip.

[0271] The memory 104 may include random access memory (RAM) or read-only memory (ROM). The memory 104 can be used to store instructions, programs, code, code sets, or instruction sets. The memory 104 may include a program storage area and a data storage area. The program storage area may store instructions for implementing an operating system, instructions for implementing at least one function, etc. The data storage area may also store data acquired during the use of the electronic device 100 (e.g., advertising data and accumulated parameter values).

[0272] The electronic device 100 may also include a network module and a screen. The network module is used to receive and transmit electromagnetic waves, converting electromagnetic waves into electrical signals, thereby enabling communication with communication networks or other devices, such as audio playback devices. The network module may include various existing circuit elements used to perform these functions, such as antennas, radio frequency transceivers, digital signal processors, encryption / decryption chips, SIM cards, memory, etc. The network module can communicate with various networks such as the Internet, corporate intranets, and wireless networks, or communicate with other devices via wireless networks. The aforementioned wireless networks may include cellular telephone networks, wireless local area networks, or metropolitan area networks. The screen can display interface content and facilitate data interaction.

[0273] The electronic device 100 may also include a network module and a screen. The network module is used to receive and transmit electromagnetic waves, converting electromagnetic waves into electrical signals, thereby enabling communication with communication networks or other devices, such as audio playback devices. The network module may include various existing circuit elements used to perform these functions, such as antennas, radio frequency transceivers, digital signal processors, encryption / decryption chips, SIM cards, memory, etc. The network module can communicate with various networks such as the Internet, corporate intranets, and wireless networks, or communicate with other devices via wireless networks. The aforementioned wireless networks may include cellular telephone networks, wireless local area networks, or metropolitan area networks. The screen can display interface content and facilitate data interaction.

[0274] In some embodiments, the electronic device 100 may further include a peripheral interface and at least one peripheral device. The processor 102, memory 104, and peripheral interface 106 can be connected via a bus or signal line. Each peripheral device can be connected to the peripheral interface via a bus, signal line, or circuit board. Specifically, the peripheral device includes at least one of the following: a radio frequency component 108, a positioning component 112, a camera 114, an audio component 116, and a display screen 118.

[0275] Peripheral interface 106 can be used to connect at least one I / O (Input / Output) related peripheral device to processor 102 and memory 104. In some embodiments, processor 102, memory 104 and peripheral interface 106 are integrated on the same chip or circuit board; in some other embodiments, any one or two of processor 102, memory 104 and peripheral interface 106 can be implemented on separate chips or circuit boards, and this application embodiment does not limit this.

[0276] The radio frequency (RF) component 108 is used to receive and transmit RF (Radio Frequency) signals, also known as electromagnetic signals. The RF component 108 communicates with communication networks and other communication devices via electromagnetic signals. The RF component 108 converts electrical signals into electromagnetic signals for transmission, or converts received electromagnetic signals back into electrical signals. Optionally, the RF component 108 includes: an antenna system, an RF transceiver, one or more amplifiers, a tuner, an oscillator, a digital signal processor, a codec chipset, a user identity module card, etc. The RF component 108 can communicate with other terminals via at least one wireless communication protocol. This wireless communication protocol includes, but is not limited to: the World Wide Web, metropolitan area networks, intranets, various generations of mobile communication networks (2G, 3G, 4G, and 5G), wireless local area networks, and / or WiFi (Wireless Fidelity) networks. In some embodiments, the RF component 108 may also include circuitry related to NFC (Near Field Communication), which is not limited in this application.

[0277] Positioning component 112 is used to locate the current geographic location of an electronic device to enable navigation or LBS (Location Based Service). Positioning component 112 can be a positioning component based on the US GPS (Global Positioning System), China's BeiDou system, or Russia's Galileo system, etc.

[0278] Camera 114 is used to capture images or videos. Optionally, camera 114 includes a front-facing camera and a rear-facing camera. Typically, the front-facing camera is located on the front panel of the electronic device 100, and the rear-facing camera is located on the back of the electronic device 100. In some embodiments, there are at least two rear-facing cameras, which are any one of a main camera, a depth-sensing camera, a wide-angle camera, and a telephoto camera, to achieve background blurring by fusion of the main camera and the depth-sensing camera, panoramic shooting by fusion of the main camera and the wide-angle camera, VR (Virtual Reality) shooting, or other fusion shooting functions. In some embodiments, camera 114 may also include a flash. The flash can be a single-color temperature flash or a dual-color temperature flash. A dual-color temperature flash refers to a combination of a warm light flash and a cool light flash, which can be used for light compensation at different color temperatures.

[0279] Audio component 116 may include a microphone and a speaker. The microphone is used to collect sound waves from the user and the environment, and convert the sound waves into electrical signals that are input to processor 102 for processing, or input to radio frequency component 108 for voice communication. For stereo acquisition or noise reduction purposes, there may be multiple microphones, each located at a different part of electronic device 100. The microphone may also be an array microphone or an omnidirectional microphone. The speaker is used to convert electrical signals from processor 102 or radio frequency component 108 into sound waves. The speaker may be a conventional diaphragm speaker or a piezoelectric ceramic speaker. When the speaker is a piezoelectric ceramic speaker, it can convert electrical signals not only into sound waves that humans can hear, but also into sound waves that humans cannot hear for purposes such as ranging. In some embodiments, audio component 114 may also include a headphone jack.

[0280] Display screen 118 is used to display a UI (User Interface). This UI may include graphics, text, icons, videos, and any combination thereof. When display screen 118 is a touch display screen, it also has the ability to collect touch signals on or above its surface. These touch signals can be input as control signals to processor 102 for processing. In this case, display screen 118 can also be used to provide virtual buttons and / or a virtual keyboard, also known as soft buttons and / or a soft keyboard. In some embodiments, there may be one display screen 118, which serves as the front panel of electronic device 100; in other embodiments, there may be at least two display screens, respectively disposed on different surfaces of electronic device 100 or in a folded design; in still other embodiments, display screen 118 may be a flexible display screen, disposed on a curved or folded surface of electronic device 100. Furthermore, display screen 118 may be configured as a non-rectangular irregular shape, i.e., a non-rectangular screen. Display screen 118 may be made of materials such as LCD (Liquid Crystal Display) or OLED (Organic Light-Emitting Diode).

[0281] This application also provides a computer-readable storage medium. This computer-readable medium stores program code that can be called by a processor to execute the methods described in the above method embodiments.

[0282] Computer-readable storage media can be electronic storage devices such as flash memory, EEPROM (Electrically Erasable Programmable Read-Only Memory), EPROM, hard disk, or ROM. Optionally, computer-readable storage media includes non-transitory computer-readable storage medium. The computer-readable storage medium has storage space for program code that performs any of the method steps described above. This program code can be read from or written to one or more computer program products. The program code can be compressed, for example, in a suitable form.

[0283] This application also provides a computer program product or computer program that includes computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform the methods described in the various optional implementations above.

[0284] In summary, the advertising conversion rate calibration method, apparatus, electronic device, and storage medium provided in this application acquire cumulative parameter values ​​of multiple sets of target advertising data within a preset time period. Each set of target advertising data has the same dimensional information and corresponds to the same dimensional combination under at least one dimensional object. The target advertising data is selected from M sets of advertising data, which also include advertising data to be deployed. Based on the advertising data to be deployed and the target dimensional combination corresponding to it, cumulative target parameter values ​​are selected from the cumulative parameter values ​​of the multiple sets of target advertising data within the preset time period. A calibration coefficient for the advertising data to be deployed is obtained based on the cumulative target parameter values. The calibration coefficient is then used to calibrate the estimated conversion rate of the advertising data to be deployed, resulting in a calibrated estimated conversion rate. This allows for the acquisition of accurate calibration coefficients based on the cumulative target parameter values ​​of target advertising data that are highly correlated with or similar to the advertising data to be deployed during the calibration process, thereby making the calibrated predicted conversion rate obtained by calibrating the predicted conversion rate using these calibration coefficients more accurate.

[0285] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims

1. A method for calibrating advertising conversion rates, characterized in that, The method includes: Obtain M pieces of advertising data and the corresponding identifier combination for each piece of advertising data. The M pieces of advertising data include advertising data to be delivered. Each piece of advertising data includes N dimensions, each dimension corresponding to a different dimension object. M is an integer greater than 1, and N is an integer greater than 1. The identifier combination consists of at least two of the following identifiers from the corresponding advertising data: advertising playback platform identifier, advertising launch information identifier, and advertising audience identifier. The dimension object is used to distinguish the data category to which each dimension of the advertising data belongs. The dimension information is the specific information included in each dimension object. The cumulative parameter values ​​related to the ad conversion rate of multiple sets of target ad data are obtained within a preset time period. Each set of target ad data has the same dimensional information and corresponds to the same identifier combination under at least one dimensional object. The target ad data is selected from M ad data. The cumulative parameter values ​​include one or more of the following: cumulative conversion value, cumulative estimated conversion rate value, cumulative estimated click-through rate value, cumulative click count value, and cumulative cost value. Based on the advertising data to be delivered and the corresponding identifier combination, select the target parameter cumulative value from the parameter cumulative values ​​corresponding to the multiple sets of target advertising data within a preset time period; The calibration coefficient of the advertising data to be delivered is obtained based on the cumulative value of the target parameter. The estimated conversion rate of the advertising data to be delivered is calibrated using the calibration coefficient to obtain the calibrated estimated conversion rate.

2. The method according to claim 1, characterized in that, Based on the advertising data to be delivered and the corresponding identifier combination, target parameter cumulative values ​​are selected from the cumulative parameter values ​​corresponding to the multiple sets of target advertising data within a preset time period, including: Based on the advertising data to be delivered, the corresponding identifier combination, and the priority order of N dimension objects, the target parameter cumulative value is selected from the parameter cumulative values ​​corresponding to the multiple sets of target advertising data within a preset time period.

3. The method according to claim 2, characterized in that, Each set of target ad data includes ad data with the same ad consumption range. The ad consumption is the cost incurred when the corresponding ad data is played on the ad playback platform. The preset duration includes a first preset duration and a second preset duration, and the first preset duration is less than the second preset duration. The cumulative parameter values ​​of multiple sets of target ad data within the preset duration are obtained, including: The cumulative values ​​of a first parameter related to the advertising conversion rate of multiple sets of target advertising data within the first preset time period are obtained, as well as the cumulative values ​​of a second parameter related to the advertising conversion rate of multiple sets of target advertising data within the second preset time period. Based on the advertising data to be delivered, the corresponding identifier combination, and the priority order of the N dimension objects, target parameter cumulative values ​​are selected from the cumulative parameter values ​​corresponding to the multiple sets of target advertising data, including: From the cumulative values ​​of the first parameters corresponding to the multiple sets of target advertising data, select the cumulative value of the first initial parameter corresponding to each dimension of the advertising data to be delivered, and the identifier combination corresponding to the first initial parameter cumulative value is the identifier combination corresponding to the advertising data to be delivered. Detect whether there is a first target consumption accumulation value among the accumulated values ​​of each of the first initial parameters, wherein the minimum value of the advertising consumption range corresponding to the first target consumption accumulation value is greater than a preset advertising consumption threshold. If it does not exist, then based on the advertising data to be delivered, the identifier combination corresponding to the advertising data to be delivered, and the priority order of the N dimension objects, the target parameter cumulative value is selected from the second parameter cumulative values ​​corresponding to the multiple sets of target advertising data respectively.

4. The method for calibrating advertising conversion rates according to claim 3, characterized in that, The step of selecting target parameter cumulative values ​​from the cumulative parameter values ​​corresponding to the multiple sets of target advertising data within a preset time period based on the advertising data to be delivered and the identifier combination corresponding to the advertising data to be delivered, further includes: If a first target parameter cumulative value exists among the cumulative values ​​of each of the first initial parameters, then the number of such first target parameter cumulative values ​​is obtained; If the first target parameter has a cumulative value of one, then the first target parameter cumulative value is used as the target parameter cumulative value. If there are at least two cumulative values ​​for the first target parameter, then a cumulative target parameter value is selected from the at least two cumulative values ​​for the first target parameter based on the dimension information corresponding to the at least two cumulative values ​​for the first target parameter, the dimension object to which the dimension information belongs, and the priority order of the N types of dimension objects.

5. The method according to claim 1, characterized in that, The cumulative values ​​of the target parameters include the cumulative value of the target reported conversions, the cumulative value of the target estimated conversion rate, the cumulative value of the target estimated click-through rate, and the cumulative value of the target clicks. The calibration coefficients for the advertising data to be deployed are obtained based on the cumulative values ​​of the target parameters, including: The initial calibration coefficients of the advertising data to be deployed are obtained by numerically calculating the cumulative value of the target reported conversion number, the cumulative value of the target estimated conversion rate, the cumulative value of the target estimated click rate, and the cumulative value of the target click number. For each of the multiple exposure stages, based on the cumulative consumption and cumulative reported conversions of the advertising data to be delivered within a preset duration of the exposure stage, the exposure stage is divided into a first exposure stage or a second exposure stage; the exposure stage refers to the stage in which the advertising data to be delivered is exposed. Based on the cumulative consumption value of the advertising data to be delivered, the consumption stage of the advertising data to be delivered is determined; the consumption stage refers to the stage in which the advertising data to be delivered generates advertising consumption, and the advertising consumption of the advertising data to be delivered is the cost generated when the advertising data to be delivered is played on the advertising playback platform. The magnification factor is obtained based on the exposure stage and the consumption stage; the magnification factor is a factor used to calibrate the initial calibration factor. The initial calibration coefficient is processed based on the amplification coefficient to obtain the calibration coefficient of the advertising data to be delivered.

6. The method according to claim 5, characterized in that, For each of the multiple exposure stages, based on the cumulative consumption and cumulative reported conversions of the advertising data to be delivered within a preset duration of that exposure stage, the exposure stage is divided into a first exposure stage or a second exposure stage, including: For each of the multiple exposure stages, if the cumulative consumption value of the advertising data to be delivered within the preset duration of the exposure stage is less than the preset cumulative consumption threshold, and the cumulative reported conversion value of the advertising data to be delivered within the preset duration of the exposure stage is less than the preset conversion threshold, the exposure stage is divided into the first exposure stage. If the cumulative consumption value of the advertising data to be delivered within the preset duration of the exposure phase is not less than the preset cumulative consumption threshold, or the cumulative reported conversion value of the advertising data to be delivered within the preset duration of the exposure phase is not less than the preset conversion threshold, then the exposure phase is divided into the second exposure phase.

7. The method according to claim 6, characterized in that, The step of obtaining the amplification factor of the initial calibration coefficient based on the exposure stage and the consumption stage includes: When the exposure phase is the first exposure phase and the consumption phase is the first consumption phase, a preset constant is obtained, which is the expansion coefficient of the initial calibration coefficient; the first consumption phase is the consumption phase in which the cumulative consumption value of the advertising data to be delivered is in the range of [0, 8*target_cpa), where target_cpa is the conversion bid of the advertising data to be delivered, and the conversion bid refers to the cost paid for each conversion of the advertising data to be delivered; When the exposure phase is the first exposure phase and the consumption phase is the second consumption phase, a deviation value of the advertising data to be delivered is obtained based on the cumulative consumption value and cumulative conversion value of the advertising data to be delivered within the preset time period, as well as the conversion bid of the advertising data to be delivered. A first deviation correction parameter is obtained based on the cumulative consumption value of the advertising data to be delivered and the first target consumption value in the consumption range corresponding to the second consumption phase. An expansion coefficient is obtained based on the first deviation correction parameter and the deviation value, wherein the maximum consumption value in the consumption range corresponding to the first consumption phase is less than the minimum consumption value in the consumption range corresponding to the second consumption phase; the second consumption phase is when the cumulative consumption value of the advertising data to be delivered is in the range [8*target_cpa, 25*target_cpa]. The consumption phase is within the range of [pa]. When the exposure phase is the first exposure phase and the consumption phase is the third consumption phase, the deviation value of the advertising data to be delivered is obtained based on the cumulative consumption value and cumulative conversion value of the advertising data to be delivered within the preset time period and the conversion bid of the advertising data to be delivered. A second deviation correction parameter is obtained based on the second target consumption value and the third consumption value in the consumption range corresponding to the third consumption phase. An expansion coefficient is obtained based on the second deviation correction parameter and the deviation value. The maximum consumption value in the consumption value range corresponding to the second consumption phase is less than the minimum consumption value in the consumption value range corresponding to the third consumption phase. The third consumption phase is the consumption phase where the cumulative consumption value of the advertising data to be delivered is greater than 25*target_cpa.

8. The method according to claim 7, characterized in that, The step of obtaining the amplification factor of the initial calibration factor based on the exposure stage and the consumption stage further includes: When the exposure stage is the second exposure stage, the smoothing coefficient of the advertising data to be delivered is obtained based on the historical cumulative conversion value of the advertising data to be delivered and the preset duration. The expansion coefficient of the advertising data to be deployed is obtained based on the historical conversion volume and historical conversion rate estimated cumulative value of the advertising data to be deployed, the smoothing coefficient, and the target parameter cumulative value.

9. The method according to claim 1, characterized in that, When there are multiple cumulative values ​​for the target parameters, before obtaining the calibration coefficient of the advertising data to be delivered based on the cumulative values ​​of the target parameters, the method further includes: The cumulative values ​​of multiple target parameters are weighted and summed to obtain the weighted summed cumulative values ​​of the target parameters.

10. The method according to any one of claims 1-9, characterized in that, The step of calibrating the estimated conversion rate of the advertisement data to be delivered using the calibration coefficient to obtain the calibrated estimated conversion rate includes: The calibration coefficient is multiplied by the estimated conversion rate of the advertising data to be delivered to obtain the calibrated estimated conversion rate.

11. The method according to any one of claims 1-9, characterized in that, The dimension objects include at least one of the following: advertiser dimension, product dimension, and advertiser dimension.

12. The method according to any one of claims 1-9, characterized in that, The identifier combination corresponding to the advertising data consists of the advertising platform identifier and the advertising launch information identifier in the advertising data.

13. A device for calibrating advertising conversion rates, characterized in that, The device includes: The first acquisition module is used to acquire M pieces of advertising data and the corresponding identifier combination for each piece of advertising data. The M pieces of advertising data include advertising data to be delivered. Each piece of advertising data includes N dimensions, each dimension corresponding to a different dimension object. M is an integer greater than 1, and N is an integer greater than 1. The identifier combination consists of at least two of the following identifiers from the corresponding advertising data: advertising playback platform identifier, advertising launch information identifier, and advertising audience identifier. The dimension object is used to distinguish the data category to which each dimension of the advertising data belongs. The dimension information is the specific information included in each dimension object. The second acquisition module is used to acquire the cumulative parameter values ​​related to the ad conversion rate of multiple sets of target ad data within a preset time period. Each set of target ad data has the same dimensional information and corresponds to the same identifier combination under at least one dimensional object. The target ad data is selected from M ad data. The cumulative parameter values ​​include one or more of the following: cumulative conversion value, cumulative estimated conversion rate value, cumulative estimated click-through rate value, cumulative click count value, and cumulative cost value. The parameter accumulation value selection module is used to select a target parameter accumulation value from the parameter accumulation values ​​corresponding to the multiple sets of target advertising data within a preset time period, based on the advertising data to be delivered and the identifier combination corresponding to the advertising data to be delivered; The calibration coefficient acquisition module is used to acquire the calibration coefficient of the advertisement data to be delivered based on the cumulative value of the target parameter. The conversion rate calibration module is used to calibrate the estimated conversion rate of the advertising data to be delivered using the calibration coefficient, so as to obtain the calibrated estimated conversion rate.

14. An electronic device, characterized in that, include: One or more processors; Memory; One or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, the one or more programs being configured to perform the method as described in any one of claims 1-12.

15. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores program code that can be invoked by a processor to execute the method as described in any one of claims 1-12.

Images