Advertisement delivery method, device, electronic device and storage medium

By identifying advertising chasing users and calculating evaluation indicators, using different evaluation indicators for different types of advertisements, the problem of improving APP retention rate in the online service industry is solved, and the retention rate is improved while taking into account the stickiness of advertising platforms. rate effect.

CN114897562BActive Publication Date: 2024-11-05BEIJING QIYI CENTURY SCI & TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202210592156.8
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2022-05-27
Publication Date
2024-11-05
Estimated Expiration
2042-05-27

AI Technical Summary

Technical Problem

In the online service industry, how to improve the retention rate of APP while taking into account the user's stickiness to the advertising platform is an urgent problem. The existing technology is difficult to effectively improve the user's stickiness and retention rate of APP.

Method used

By identifying the users of advertising chasing, calculating the evaluation indicators of target advertisements and candidate advertisements, including the first-class evaluation indicators and the second-class evaluation indicators, the advertisements are evaluated and sorted based on these indicators, and thus selecting suitable advertisements for delivery, targeting advertising chasing Investment users and non-tracking investors use different evaluation indicators respectively to improve retention rate.

Benefits of technology

It has achieved the improvement of APP retention rate of online service customers while taking into account the user's stickiness to the advertising platform. By using different evaluation indicators for different types of advertisements, the effectiveness of advertising delivery is improved.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN114897562B_ABST
    Figure CN114897562B_ABST
Patent Text Reader

Abstract

The embodiments of the present invention provide an advertisement delivery method, device, electronic device and storage medium, and relate to the field of computer technology. The method comprises: identifying whether the target user to be delivered an advertisement is an advertisement follow-up user; if the identification result is yes, calculating the first type of evaluation index of the target advertisement, and calculating the second type of evaluation index of other candidate advertisements other than the target advertisement; based on the calculated first type of evaluation index and second type of evaluation index, selecting the advertisement to be delivered from the target advertisement and other candidate advertisements; delivering the advertisement to be delivered to the target user. Through this solution, it is possible to improve the secondary retention rate of the APP of the network service customer while taking into account the user's stickiness to the advertising platform.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to an advertisement delivery method, device, electronic equipment and storage medium. Background Art

[0002] The Internet service industry refers to the industry that provides APP (Application) services through the Internet. For Internet service customers, continuously finding more new users for their own APP is a necessary means to ensure the long-term development of APP. Among them, the behavior of finding new users for APP is referred to as Internet service acquisition. In Internet service acquisition, Internet advertising is an important channel, that is, by placing advertisements on the Internet to attract new users for APP. For Internet service customers, after acquiring new users, they need to measure the quality of users to ensure their own ROI (return on investment).

[0003] For network service customers, a common indicator for measuring user quality is the second retention rate. The higher the second retention rate, the higher the user stickiness to the APP, which is more beneficial to network service customers. Among them, the second retention rate refers to the proportion of users who retain for the second time the next day among the users activated on the same day, that is, the number of users who retain for the second time / the number of activated users. The so-called second retention refers to the number of users who open the APP again the next day after downloading it.

[0004] During the advertising process, how to improve the retention rate of the online service customers' APP while taking into account the user's stickiness to the advertising platform is an urgent problem to be solved. Summary of the invention

[0005] The purpose of the embodiments of the present invention is to provide an advertisement delivery method, device, electronic device and storage medium, so as to improve the secondary retention rate of the APP of the network service customer while taking into account the user's stickiness to the advertising platform. The specific technical solution is as follows:

[0006] In a first aspect, an embodiment of the present invention provides an advertisement delivery method, the method comprising:

[0007] Identify whether the target user to be advertised is a follow-up user; wherein the follow-up user is a user who has visited the advertisement of the target APP in the last specified period; the target APP is the APP for which the retention rate is to be improved;

[0008] If the recognition result is yes, a first type of evaluation index of the target advertisement is calculated, and a second type of evaluation index of other candidate advertisements other than the target advertisement is calculated; wherein the target advertisement is an advertisement of the target APP, the first type of evaluation index is used to characterize the interest level of the target user in the advertisement re-delivered to the target APP, and the second type of evaluation index is used to characterize the interest level of the target user in other candidate advertisements other than the target advertisement;

[0009] Based on the calculated first evaluation index and second evaluation index, selecting an advertisement to be placed from the target advertisement and other candidate advertisements;

[0010] The advertisement to be delivered is delivered to the target user.

[0011] Optionally, the target advertisement is a historical advertisement of the target APP, or each advertisement of the target APP, and each advertisement includes all or part of the advertisements of the target APP;

[0012] The historical advertisements are advertisements visited by the target user during the last specified time period.

[0013] Optionally, the step of identifying whether the target user to which the advertisement is to be delivered is a follow-up advertisement delivery user includes:

[0014] Detecting whether there is a target history record of the target user, and if so, identifying the target user as an advertisement follow-up user;

[0015] The target history record is a record of visits to advertisements of a target APP with a retention assessment requirement within a previously specified period of time.

[0016] Optionally, the calculating of the first type of evaluation index of the target advertisement includes:

[0017] Calculate a first type of evaluation index of the target advertisement based on a first estimated click-through rate, a click-to-buy rate, and a click-to-buy bid of the target advertisement;

[0018] The first estimated click rate is the probability that the target user clicks on the target advertisement after the target advertisement is delivered;

[0019] The click-through rate is the probability of a successful follow-up after the target user clicks on the target advertisement, where a successful follow-up means clicking into the target APP;

[0020] The follow-up bid is the bid that the advertiser of the target APP needs to give when placing the advertisement of the target APP again.

[0021] Optionally, the calculating of the first type of evaluation index of the target advertisement based on the first estimated click-through rate, click-to-buy rate, and click-to-buy bid of the target advertisement includes:

[0022] Based on the first estimated click-through rate, click-to-buy rate, and buy-to-buy bid of the target advertisement, a first type of evaluation index of the target advertisement is calculated using a predetermined buy-to-buy formula;

[0023] Wherein, the predetermined follow-up investment formula includes:

[0024] ECPM 追投 =price 追投 *pCTR*pCRR;

[0025] Among them, ECPM 追投 For the first type of evaluation indicators, price 追投 is the follow-up bid, pCTR is the first estimated click-through rate, and pCRR is the click-through rate.

[0026] Optionally, the follow-up bid is calculated using the following formula:

[0027] price 追投 =price*M;

[0028] Wherein, price is the bid of the target advertisement for the second retention conversion, M is the price correction factor and M=1-pDCVR, and pDCVR is the first deep conversion rate;

[0029] The first deep conversion rate is the probability that the target user completes the second retention after downloading the target APP.

[0030] Optionally, the method for determining the click-through rate includes:

[0031] Obtaining the specified portrait information of the target user and the specified attribute information of the target APP;

[0032] The click-through rate is calculated based on the designated portrait information and the designated attribute information.

[0033] Optionally, the calculating of the second type of evaluation indicators of other candidate advertisements except the target advertisement includes:

[0034] Calculate the second type of evaluation index of the other candidate advertisements based on the conversion bids, second estimated click-through rates, estimated conversion rates, and second deep conversion rates of the other candidate advertisements except the target advertisement;

[0035] The conversion bid is a bid of the other candidate advertisements for achieving a specified conversion;

[0036] The second estimated click-through rate is the probability that the target user clicks on the other candidate advertisement after the other candidate advertisement is delivered;

[0037] The estimated conversion rate is the probability that the target user downloads the APP to which the other candidate advertisement belongs after visiting the other candidate advertisement;

[0038] The second deep conversion rate is the probability that the specified conversion will occur after the target user downloads the APP to which the other candidate advertisements belong.

[0039] Optionally, selecting an advertisement to be delivered from the target advertisement and the other candidate advertisements based on the calculated first evaluation index and the second evaluation index includes:

[0040] Based on the calculated first evaluation index and second evaluation index, the target advertisement and the other candidate advertisements are evaluated and ranked from high to low;

[0041] Based on the ranking result, a specified number of advertisements with high rankings are selected from the target advertisement and the other candidate advertisements to obtain the advertisements to be delivered.

[0042] In a second aspect, an embodiment of the present invention provides an advertisement delivery device, including:

[0043] An identification module is used to identify whether the target user to be advertised is a follow-up advertising user; wherein the follow-up advertising user is a user who has visited the advertisement of the target APP within the last specified period; the target APP is the APP for which the retention rate is to be improved;

[0044] A calculation module, for calculating, if the identification result is yes, a first type of evaluation index of the target advertisement, and a second type of evaluation index of other candidate advertisements other than the target advertisement; wherein the target advertisement is an advertisement of the target APP, the first type of evaluation index is used to characterize the interest level of the target user in the advertisement re-delivered to the target APP, and the second type of evaluation index is used to characterize the interest level of the target user in other candidate advertisements other than the target advertisement;

[0045] A selection module, configured to select an advertisement to be placed from the target advertisement and other candidate advertisements based on the first evaluation index and the second evaluation index obtained by calculation;

[0046] A delivery module is used to deliver the advertisement to be delivered to the target user.

[0047] Optionally, the target advertisement is a historical advertisement of the target APP, or each advertisement of the target APP, and each advertisement includes all or part of the advertisements of the target APP;

[0048] The historical advertisements are advertisements visited by the target user during the last specified time period.

[0049] Optionally, the identification module is specifically used for:

[0050] Detecting whether there is a target history record of the target user, and if so, identifying the target user as an advertisement follow-up user;

[0051] The target history record is a record of visits to advertisements of a target APP with a retention assessment requirement within a previously specified period of time.

[0052] Optionally, the calculation module calculates the first type of evaluation index of the target advertisement, including:

[0053] Calculate a first type of evaluation index of the target advertisement based on a first estimated click-through rate, a click-to-buy rate, and a click-to-buy bid of the target advertisement;

[0054] The first estimated click rate is the probability that the target user clicks on the target advertisement after the target advertisement is delivered;

[0055] The click-through rate is the probability of a successful follow-up after the target user clicks on the target advertisement, where a successful follow-up means clicking into the target APP;

[0056] The follow-up bid is the bid that the advertiser of the target APP needs to give when placing the advertisement of the target APP again.

[0057] Optionally, the calculation module calculates a first type of evaluation index of the target advertisement based on a first estimated click-through rate, a click-to-buy rate, and a click-to-buy bid of the target advertisement, including:

[0058] Based on the first estimated click-through rate, click-to-buy rate, and buy-to-buy bid of the target advertisement, a first type of evaluation index of the target advertisement is calculated using a predetermined buy-to-buy formula;

[0059] Wherein, the predetermined follow-up investment formula includes:

[0060] ECPM 追投 =price 追投 *pCTR*pCRR;

[0061] Among them, ECPM 追投 For the first type of evaluation indicators, price 追投 is the follow-up bid, pCTR is the first estimated click-through rate, and pCRR is the click-through rate.

[0062] Optionally, the follow-up bid is calculated using the following formula:

[0063] price 追投 =price*M;

[0064] Wherein, price is the bid of the target advertisement for the second retention conversion, M is the price correction factor and M=1-pDCVR, and pDCVR is the first deep conversion rate;

[0065] The first deep conversion rate is the probability that the target user completes the second retention after downloading the target APP.

[0066] Optionally, the method for determining the click-through rate includes:

[0067] Obtaining the specified portrait information of the target user and the specified attribute information of the target APP;

[0068] The click-through rate is calculated based on the designated portrait information and the designated attribute information.

[0069] Optionally, the calculation module calculates the second type of evaluation indicators of other candidate advertisements except the target advertisement, including:

[0070] Calculate the second type of evaluation index of the other candidate advertisements based on the conversion bids, second estimated click-through rates, estimated conversion rates, and second deep conversion rates of the other candidate advertisements except the target advertisement;

[0071] The conversion bid is a bid of the other candidate advertisements for achieving a specified conversion;

[0072] The second estimated click-through rate is the probability that the target user clicks on the other candidate advertisement after the other candidate advertisement is delivered;

[0073] The estimated conversion rate is the probability that the target user downloads the APP to which the other candidate advertisement belongs after visiting the other candidate advertisement;

[0074] The second deep conversion rate is the probability that the specified conversion will occur after the target user downloads the APP to which the other candidate advertisements belong.

[0075] Optionally, the selection module is specifically used to:

[0076] Based on the calculated first evaluation index and second evaluation index, the target advertisement and the other candidate advertisements are evaluated and ranked from high to low;

[0077] Based on the ranking result, a specified number of advertisements with high rankings are selected from the target advertisement and the other candidate advertisements to obtain the advertisements to be delivered.

[0078] In a third aspect, an embodiment of the present invention provides an electronic device, including a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other via the communication bus;

[0079] Memory, used to store computer programs;

[0080] The processor is used to implement any advertisement delivery method when executing the program stored in the memory.

[0081] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, any advertisement delivery method is implemented.

[0082] In a fifth aspect, an embodiment of the present invention further provides a computer program product comprising instructions, which, when executed on a computer, enables the computer to execute any of the above-mentioned advertising delivery methods.

[0083] Beneficial effects of the embodiments of the present invention:

[0084] In this scheme, when delivering advertisements to target users who are advertising chasing users, the first type of evaluation index is calculated for the target advertisement, and the second type of evaluation index is calculated for other candidate advertisements, and then the advertisements are evaluated and ranked based on the first type of evaluation index and the second type of evaluation index, so as to select the advertisements for delivery to the target users. Since the advertisements of the target APP can arouse the interest of the target users in the target APP, that is, there is a possibility of improving the secondary retention rate of the target APP, therefore, this scheme uses different types of evaluation indexes as considerations in the delivery process for different types of advertisements: advertisements that have an impact on the secondary retention rate and advertisements that do not have an impact, that is, advertisements related to the target APP and advertisements that are not related. It can be seen that through this scheme, for advertising chasing users, when delivering advertisements, different evaluation indicators are used as considerations in the delivery process for different types of advertisements. Therefore, the secondary retention rate of the APP of the network service customer can be improved while taking into account the user's stickiness to the advertising platform.

[0085] Of course, it is not necessary to achieve all of the advantages described above at the same time to implement any product or method of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0086] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art are briefly introduced below.

[0087] Figure 1A flowchart of an advertisement delivery method provided by an embodiment of the present invention;

[0088] Figure 2 Another flow chart of an advertisement delivery method provided by an embodiment of the present invention;

[0089] FIG3( a ) is a schematic diagram of a flow chart of an advertisement delivery embodiment provided by an embodiment of the present invention;

[0090] FIG3( b ) is another schematic diagram of an advertisement delivery embodiment provided by an embodiment of the present invention;

[0091] Figure 4 A schematic diagram of the structure of an advertisement delivery device provided by an embodiment of the present invention;

[0092] Figure 5 A schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0093] The technical solutions in the embodiments of the present invention will be described below in conjunction with the accompanying drawings in the embodiments of the present invention.

[0094] The embodiments of the present invention provide an advertisement delivery method, device, electronic device and storage medium, so as to improve the secondary retention rate of the APP of the network service customer while taking into account the user's stickiness to the advertisement platform during advertisement delivery.

[0095] An advertisement delivery method provided in an embodiment of the present invention can be applied to an advertisement delivery device, wherein the advertisement delivery device is a device deployed with an advertisement platform, and the advertisement platform is a functional software for implementing advertisement delivery. In practical applications, the advertisement delivery device can be any type of device capable of deploying an advertisement delivery program, such as a server or a terminal device, which is reasonable.

[0096] It should be noted that, in a specific application, the advertising platform can be associated with the advertising playback end, so that after the advertising platform determines the advertisement to be delivered, the selected advertisement to be delivered is delivered to the user, which can specifically include: displaying the selected advertisement to the user through the advertising playback end, thereby achieving advertising delivery. Exemplarily, for a video service provider, the video service provider can have an advertising platform and a video client, wherein the advertising platform communicates with the video client and displays advertisements to the user through the video client. Of course, in another optional implementation method, the advertising platform has an advertisement playback function. At this time, after the advertising platform determines the advertisement to be delivered, the selected advertisement to be delivered is delivered to the user, which can specifically include: displaying the selected advertisement to be delivered to the user.

[0097] In order to solve the problems of the related art, an advertisement delivery method provided by an embodiment of the present invention may include the following steps:

[0098] Identify whether the target user to be advertised is a follow-up user; wherein the follow-up user is a user who has visited the advertisement of the target APP in the last specified period; the target APP is the APP for which the retention rate is to be improved;

[0099] If the recognition result is yes, a first type of evaluation index of the target advertisement is calculated, and a second type of evaluation index of other candidate advertisements other than the target advertisement is calculated; wherein the target advertisement is an advertisement of the target APP, the first type of evaluation index is used to characterize the interest level of the target user in the advertisement re-delivered to the target APP, and the second type of evaluation index is used to characterize the interest level of the target user in other candidate advertisements other than the target advertisement;

[0100] Based on the calculated first evaluation index and second evaluation index, selecting an advertisement to be placed from the target advertisement and other candidate advertisements;

[0101] The advertisement to be delivered is delivered to the target user.

[0102] In this scheme, when delivering advertisements to target users who are advertising chasing users, the first type of evaluation index is calculated for the target advertisement, and the second type of evaluation index is calculated for other candidate advertisements, and then the advertisements are evaluated and ranked based on the first type of evaluation index and the second type of evaluation index, so as to select the advertisements for delivery to the target users. Since the advertisements of the target APP can arouse the interest of the target users in the target APP, that is, there is a possibility of improving the secondary retention rate of the target APP, therefore, this scheme uses different types of evaluation indexes as considerations in the delivery process for different types of advertisements: advertisements that have an impact on the secondary retention rate and advertisements that do not have an impact, that is, advertisements related to the target APP and advertisements that are not related. It can be seen that through this scheme, for advertising chasing users, when delivering advertisements, different evaluation indicators are used as considerations in the delivery process for different types of advertisements. Therefore, the secondary retention rate of the APP of the network service customer can be improved while taking into account the user's stickiness to the advertising platform.

[0103] In the following, an advertisement delivery method provided by an embodiment of the present invention is described in detail with reference to the accompanying drawings.

[0104] Figure 1 A flowchart of an advertisement delivery method provided by an embodiment of the present invention is shown in FIG. Figure 1 As shown, the method may include the following steps S101-S104:

[0105] S101: Identify whether the target user to be advertised is a follow-up advertisement user;

[0106] Among them, the advertising follow-up users are users who have visited the advertisements of the target APP within the last specified time period; the target APP is the APP for which the second-time retention rate is to be improved. It should be noted that if an APP uses the second-time retention as the delivery purpose, then the APP is the target APP for which the second-time retention rate is to be improved. In addition, the so-called last specified time period is a time period set based on the meaning represented by the second-time retention. For example, since the second-time retention refers to the user opening the APP again the next day after downloading the APP, the last specified time period can be the previous day, but of course it is not limited to this.

[0107] For any target user to be advertised, if the target user has visited the advertisement of the target APP in the previous specified period, that is, activated, then, if it is possible to place advertisements related to the target APP, that is, to follow up the advertisements of the target APP, it will help to awaken the target user's interest in the target APP, thereby improving the secondary retention rate of the target APP. For example: Assuming that a news APP has a secondary retention assessment requirement, that is, whether the secondary retention occurs is the purpose of the advertisement, then when it is necessary to place advertisements for the target user, if it is detected that the target user has visited the news APP the day before, then when it is necessary to place advertisements for the target user, if it is possible to place advertisements related to the news APP, that is, to follow up the advertisements of the news APP, it will help to improve the secondary retention rate of the target APP.

[0108] Based on the above ideas, in order to solve the problems existing in related technologies, after determining the target users to whom the advertisements are to be delivered, instead of directly analyzing each candidate advertisement in the advertisement library in the same way, it is necessary to identify whether the target users to whom the advertisements are to be delivered are the users to whom the advertisements are to be delivered. Then, different processing can be performed according to different identification results.

[0109] It is understandable that the target user is the user to whom advertisements are to be delivered. In different application scenarios, the method of determining the target user may be different, and the embodiments of the present invention do not limit this. Optionally, the target user may be determined by the advertisement playback end. For example, if a video client is used as the advertisement playback end, then if a user enters the homepage of the video client, or sends a video access instruction to the video client, the user may be the target user to be delivered advertisements. Then, the video client may send a request to the advertisement platform to deliver advertisements to the target user. Accordingly, the advertisement platform may determine the target user to be delivered advertisements.

[0110] In addition, it is understandable that there are multiple ways to identify whether the target user to which the advertisement is to be delivered is a user to whom the advertisement is to be delivered.

[0111] Exemplarily, in one implementation, identifying whether the target user to be placed an advertisement is an ad follow-up user may include: detecting whether there is a target history record of the target user, and if so, identifying the target user as an ad follow-up user; wherein the target history record is a record of visits to advertisements of a target APP with a secondary retention assessment requirement within the last specified period of time. For this implementation, if any user clicks on an advertisement of a target APP with a secondary retention assessment requirement, a visit record may be formed; thus, after determining the target user to be placed an advertisement, it may be detected whether there is a target history record of the target user, thereby identifying whether the target user is an ad follow-up user.

[0112] For example, the access record may record information such as user name, click time, name of the clicked target APP, etc. The above method of identifying whether the target user to be advertised is an ad follow-up user is only an example and should not constitute a limitation on the embodiments of the present invention.

[0113] S102: If the identification result is yes, calculating the first type of evaluation index of the target advertisement, and calculating the second type of evaluation index of other candidate advertisements except the target advertisement;

[0114] Among them, the target advertisement is the advertisement of the target APP, the first type of evaluation indicators is used to characterize the target user's interest in the re-delivered advertisement of the target APP, and the second type of evaluation indicators is used to characterize the target user's interest in other candidate advertisements except the target advertisement.

[0115] Since during advertisement delivery, the candidate advertisements in the advertisement library include two different types of advertisements: target advertisements, and other candidate advertisements except the target advertisements, the evaluation indicators of the two types of advertisements can be calculated based on this: the first type of evaluation indicators, and the second type of evaluation indicators. In this way, for advertisements that have an impact on the second retention rate and advertisements that do not have an impact, that is, advertisements related to the target APP and unrelated advertisements, the user's interest in different types of advertisements can be used as an evaluation indicator for the delivery process. Compared with the prior art that calculates the user's interest in all candidate advertisements in the same way, this solution can improve the APP second retention rate of network service customers while taking into account the user's stickiness to the advertising platform.

[0116] For example, in one implementation, the same ad can be re-invested for the target APP. In this case, the target ad is the historical ad of the target APP, which is the ad visited by the target user in the last specified period. The so-called re-investing for the same ad means that the ad seen by the user when activated the day before is the same as the ad re-invested the next day, that is, if the user is activated under ad A, then ad A will also be exposed to the user the next day for re-investing.

[0117] Exemplarily, in another implementation, cross-advertising follow-up can be performed on the target APP. In this case, the target advertisements are the advertisements of the target APP, and each advertisement includes all or part of the advertisements of the target APP. Among them, the so-called cross-advertising follow-up means that the advertisement seen by the user when activated the day before and the advertisement followed up the next day can be the same, or they can be different advertisements under the target APP, that is, the user is activated under advertisement A, and the user can be exposed to advertisement A or advertisement B the next day for follow-up investment. Advertisement B and advertisement A are advertisements for promoting the same APP. Among them, the premise of cross-advertising follow-up investment is that the activation advertisement and the follow-up advertisement are promoting the same APP. In this way, it can be guaranteed that the material content of the two advertisements is similar, which can ensure that the user's interest in a specific APP is awakened. Moreover, compared with the same advertisement follow-up, the cross-advertisement follow-up can significantly improve the coverage of the follow-up users, thereby improving the final follow-up effect. Moreover, the inventors have found that the cross-advertisement follow-up can increase the follow-up user coverage by 144% and the secondary retention rate by 16.51%. The follow-up user coverage is also the success rate of the follow-up. For example, if ten users have activated an APP, then within a specified time period with a secondary retention requirement, five users click to enter the APP again, and the follow-up user coverage under the APP is 50%. It should be noted that the various advertisements of the target APP can be all advertisements under the target APP, or part of the advertisements under the target APP; the part of the advertisements under the target APP can include historical advertisements or not, and the various advertisements of the target APP are not limited here.

[0118] Optionally, in one implementation, calculating the first type of evaluation index of the target advertisement may include: calculating the first type of evaluation index of the target advertisement based on a first estimated click-through rate, a click-to-buy rate, and a click-to-buy bid of the target advertisement;

[0119] The first estimated click rate is the probability that the target user clicks on the target advertisement after the target advertisement is delivered;

[0120] The click-through rate is the probability of a successful follow-up after the target user clicks on the target ad. A successful follow-up means clicking into the target APP.

[0121] The follow-up bid is the bid that the advertiser of the target APP needs to give when placing the advertisement of the target APP again.

[0122] In this implementation, when calculating the first type of evaluation indicators, the click-through rate and bid price related to ad follow-up are considered, which is different from the analysis of the user's interest in other candidate ads. In this way, it is more in line with the actual effect of the target ad on the target user. Among them, the first type of evaluation indicators can be used as an improved ECPM (Effective Cost Per Mille, revenue per thousand impressions: advertising revenue obtained for every thousand impressions).

[0123] Exemplary methods for determining the click-through rate include:

[0124] Obtain the specified profile information of the target user and the specified attribute information of the target APP;

[0125] The click-through rate is calculated based on the designated portrait information and the designated attribute information.

[0126] Exemplarily, the designated portrait information may include one or more of the target user's age, interests, and region, etc.; the designated attribute information may include the APP type, but is of course not limited to this. Among them, based on the designated portrait information and the designated attribute information, there may be multiple ways to calculate the click-through rate, for example: the designated portrait information of the target user and the designated attribute information of the target APP are used as input content, and are input into a pre-trained neural network model for estimating the click-through rate, thereby obtaining the click-through rate; wherein the neural network model for estimating the click-through rate is a model pre-trained using sample data and label information, wherein the sample data is the designated portrait information of the sample user and the designated attribute information of the sample APP, and the label information represents: when the sample user is following up with an advertisement of the sample APP, whether the follow-up is successful after the sample user clicks on the advertisement being followed up.

[0127] Furthermore, any method of calculating the first estimated click rate can be applied to the embodiment of the present invention. For example, the first estimated click rate can be calculated by using a pre-trained neural network model for identifying the estimated click rate.

[0128] Optionally, in one implementation, calculating the second type of evaluation indicators for candidate advertisements other than the target advertisement may include: calculating the second type of evaluation indicators for candidate advertisements other than the target advertisement based on the conversion bids, second estimated click-through rates, estimated conversion rates, and second deep conversion rates of candidate advertisements other than the target advertisement.

[0129] In this method, since other candidate ads are unrelated to the target APP, when analyzing the user's interest in other candidate ads, indicator information related to conventional advertising can be considered: the conversion bid of other candidate ads, the second estimated click-through rate, the estimated conversion rate and the second deep conversion rate.

[0130] The conversion bid is the bid of the other candidate ads for achieving a specified conversion; that is, the bid given by the advertiser for the candidate ads to achieve a specified conversion, that is, how much the advertiser is willing to pay for a conversion, and the specified conversion may include but is not limited to: activation, or a second retention, etc. For example: the purpose of an advertisement is a second retention, that is, a user downloads the APP of the advertisement and completes the second retention on the next day. If the conversion bid is 10, it means that the advertiser is willing to spend 10 yuan to buy a new user who retains the second time.

[0131] The second estimated click-through rate is the probability that the target user clicks on the other candidate advertisements after the other candidate advertisements are delivered; the calculation method of the second estimated click-through rate is similar to the calculation method of the first estimated click-through rate mentioned above.

[0132] The estimated conversion rate is the probability that the target user will download the APP to which the other candidate advertisement belongs after visiting the other candidate advertisement; for example, if the estimated conversion rate is 50%, it means that after the user clicks the advertisement, there is a 50% probability that the user will download the APP and activate it. The estimated conversion rate can be calculated by a pre-trained neural network model for calculating the estimated conversion rate, but is not limited to this.

[0133] The second deep conversion rate is the probability that the target user will have the specified conversion after downloading the APP to which the other candidate advertisements belong. If the specified conversion is a second retention, then the second deep conversion rate is: how likely it is that the user will have a second retention after downloading the APP; for example, if the second deep conversion rate is 50%, it means that after downloading the APP, there is a 50% probability that the user will open the APP again the next day and complete the second retention. The second deep conversion rate can be calculated using a pre-trained neural network model for calculating deep conversion rates, but is of course not limited to this.

[0134] In order to have a clear layout and a clear plan, the specific implementation method of calculating the first type of evaluation indicators of the target advertisement based on the first estimated click-through rate, click-to-follow-up rate and follow-up bid of the target advertisement, and the specific implementation method of calculating the second type of evaluation indicators of other candidate advertisements based on the conversion bids, second estimated click-through rate, estimated conversion rate and second deep conversion rate of other candidate advertisements other than the target advertisement will be explained through other embodiments later.

[0135] S103: Selecting an advertisement to be placed from the target advertisement and other candidate advertisements based on the first evaluation index and the second evaluation index obtained by calculation;

[0136] In this embodiment, there are multiple ways to select the advertisement to be placed from the target advertisement and other candidate advertisements based on the calculated first type of evaluation indicators and the second type of evaluation indicators.

[0137] Exemplarily, in one implementation, based on the calculated first evaluation index and the second evaluation index, selecting the advertisement to be placed from the target advertisement and other candidate advertisements may include: based on the calculated first evaluation index and the second evaluation index, evaluating and ranking the target advertisement and other candidate advertisements from high to low; based on the ranking result, selecting a specified number of advertisements with high rankings from the target advertisement and other candidate advertisements to obtain the advertisement to be placed. The specified number may be one or more, and the specific number value may be set according to actual conditions.

[0138] Exemplarily, in another implementation, selecting advertisements to be placed from target advertisements and other candidate advertisements based on the calculated first and second evaluation indicators may include: selecting evaluation indicators that exceed a predetermined threshold from the calculated first and second evaluation indicators; randomly selecting a specified number of evaluation indicators from the evaluation indicators that exceed the predetermined threshold, and using advertisements corresponding to the specified number of evaluation indicators as the advertisements to be placed.

[0139] The above-mentioned selection of advertisements to be placed from target advertisements and other candidate advertisements based on the calculation of the first type of evaluation indicators and the second type of evaluation indicators is merely an example and should not constitute a limitation on the embodiments of the present invention.

[0140] S104: delivering the advertisement to be delivered to the target user;

[0141] Wherein, delivering the selected advertisement to be delivered to the target user means exposing the advertisement to be delivered to the target user. Exemplarily, in one implementation, delivering the selected advertisement to be delivered to the target user may include: displaying the advertisement to be delivered to the target user through an advertisement playback terminal that is in communication with the advertisement platform.

[0142] In addition, after delivering the advertisement to the target user, the target history record of the ad follow-up user can be cleared. The target history record is the visit record of the advertisement for the target APP with the second retention assessment requirement in the previous specified period, and it will be recorded again when the user visits next time. In this way, it can be ensured that each time the advertisement for the target APP with the second retention assessment requirement is delivered, the target user of the advertisement to be delivered can be accurately detected as the ad follow-up user.

[0143] It should be noted that if it is identified that the target user is not an ad-targeting user, the second evaluation index of each candidate ad can be calculated, so that based on each second evaluation index, the ad to be delivered is selected from each candidate ad, and the ad to be delivered is delivered to the target user. Exemplarily, when selecting the ad to be delivered, the candidate ads can be evaluated and ranked from high to low based on each second evaluation index; based on the ranking result, a specified number of ads with high rankings are selected from each candidate ad to obtain the ad to be delivered.

[0144] In this scheme, when delivering advertisements to target users who are advertising chasing users, the first type of evaluation index is calculated for the target advertisement, and the second type of evaluation index is calculated for other candidate advertisements, and then the advertisements are evaluated and ranked based on the first type of evaluation index and the second type of evaluation index, so as to select the advertisements for delivery to the target users. Since the advertisements of the target APP can arouse the interest of the target users in the target APP, that is, there is a possibility of improving the secondary retention rate of the target APP, therefore, this scheme uses different types of evaluation indexes as considerations in the delivery process for different types of advertisements: advertisements that have an impact on the secondary retention rate and advertisements that do not have an impact, that is, advertisements related to the target APP and advertisements that are not related. It can be seen that through this scheme, for advertising chasing users, when delivering advertisements, different evaluation indicators are used as considerations in the delivery process for different types of advertisements. Therefore, the secondary retention rate of the APP of the network service customer can be improved while taking into account the user's stickiness to the advertising platform.

[0145] like Figure 2 As shown, an advertisement delivery method provided by an embodiment of the present invention may include the following steps:

[0146] S201: Identify whether the target user to be advertised is a follow-up advertisement user;

[0147] S202: If the identification result is yes, based on the first estimated click-through rate, click-to-buy rate, and buy-to-buy bid of the target advertisement, a predetermined buy-to-buy formula is used to calculate a first evaluation index of the target advertisement, and a second evaluation index of other candidate advertisements other than the target advertisement;

[0148] In one implementation, the predetermined follow-up investment formula includes: ECPM 追投 =price 追投 *pCTR*pCRR; where ECPM 追投 is the first type of evaluation indicator, price 追投 For follow-up bidding, pCTR is the first estimated click-through rate, and pCRR is the click-through rate.

[0149] Exemplarily, the chasing bid is calculated by the following formula:

[0150] price 追投 =price*M; wherein price is the bid of the target advertisement for the second retention conversion, M is the price correction coefficient and M=1-pDCVR, pDCVR is the first deep conversion rate; the first deep conversion rate is the probability that the target user completes the second retention after downloading the target APP.

[0151] That is, ECPM 追投 = price*(1-pDCVR)*pCTR*pCRR. This calculation method can theoretically guarantee the best follow-up bid. That is, if the ECPM follow-up bid is evaluated, the advertiser will not bid too high and exceed the cost; nor will the bid be too low and the traffic will be missed.

[0152] To help you understand how to calculate the bid, here is an example:

[0153] For example, an advertiser spends 10 yuan to buy a second retention. If the first deep conversion rate = 40% (that is, there is a 40% probability of a second retention after user activation), then when activation occurs, the advertiser can be charged 10*40% = 4 yuan; if the first deep conversion rate = 100% (that is, after user activation, a second retention will definitely occur), then when activation occurs, the advertiser can be charged 10*100% = 10 yuan.

[0154] When we follow up on the second day, if we succeed in the follow-up, that is, the user has completed the second retention, we can charge the advertiser a price that satisfies: the price collected at the time of activation + the price collected at the time of follow-up = the price of the second retention. For example, the advertiser spends 10 yuan to buy a second retention. If the first deep conversion rate = 40%, as described in the previous paragraph, on the day of activation, 10*40% = 4 yuan has been collected; if the activation is successful the next day, then the advertiser can be charged 10*(1-40%) = 6 yuan; if the first deep conversion rate = 100% (that is, after the user is activated, the second retention will definitely occur), then when the activation occurs, we have already collected 10 yuan, and even if the follow-up is successful the next day, we cannot charge the advertiser again. In summary, when following up, the customer's follow-up bid is price*(1-pDCVR), price is the bid for the second retention conversion, and pDCVR is the first deep conversion rate.

[0155] The second evaluation index of the candidate ads other than the target ad can be calculated based on the conversion bid, the second estimated click-through rate, the estimated conversion rate, and the second deep conversion rate of the candidate ads other than the target ad. 投放=price*pCTR*pCVR*pDCVR, price is the conversion bid, pCTR is the second estimated click-through rate, pCVR is the estimated conversion rate, and pDCVR is the second deep conversion rate.

[0156] To facilitate understanding of the calculation method of the second type of evaluation indicators, the following example is used to illustrate:

[0157] The conversion bid refers to the cost that the advertiser is willing to pay for a conversion. For example, the purpose of an APP advertisement is to retain users for the second time, that is, a user downloads the APP and completes the second retention on the next day. If price = 10, it means that the advertiser is willing to spend 10 yuan to buy a second-retained user. The second is the estimated click-through rate, which is the platform's estimate of the probability that a user will click. The estimated conversion rate refers to the platform's estimate of the probability that a user will convert after clicking. For example, pCVR = 50%, which means that after clicking the ad, there is a 50% probability that the user will download the APP to which the ad belongs and activate it. The second is the deep conversion rate, which is the probability that the user will have a deep conversion after downloading the APP. For example, pDCVR = 50%, which means that after downloading the APP, there is a 50% probability that the user will open the APP and complete the second retention during the time period when the second-retention demand is assessed.

[0158] S203: Selecting an advertisement to be placed from the target advertisement and other candidate advertisements based on the first evaluation index and the second evaluation index obtained by calculation;

[0159] S204: delivering the advertisement to be delivered to the target user;

[0160] In this embodiment, S201, S203, and S204 are the same as S101, S103, and S104 in the above embodiment, and are not described in detail here.

[0161] In this scheme, when delivering advertisements to target users who are advertising chasing users, the first type of evaluation index is calculated for the target advertisement, and the second type of evaluation index is calculated for other candidate advertisements, and then the advertisements are evaluated and ranked based on the first type of evaluation index and the second type of evaluation index, so as to select the advertisements for delivery to the target users. Since the advertisements of the target APP can arouse the interest of the target users in the target APP, that is, there is a possibility of improving the secondary retention rate of the target APP, therefore, this scheme uses different types of evaluation indexes as considerations in the delivery process for different types of advertisements: advertisements that have an impact on the secondary retention rate and advertisements that do not have an impact, that is, advertisements related to the target APP and advertisements that are not related. It can be seen that through this scheme, for advertising chasing users, when delivering advertisements, different evaluation indicators are used as considerations in the delivery process for different types of advertisements. Therefore, the secondary retention rate of the APP of the network service customer can be improved while taking into account the user's stickiness to the advertising platform. Moreover, through the above-mentioned calculation method of the first type of evaluation indicators, the best follow-up bid can be guaranteed in theory. That is, if the evaluation is performed according to the ECPM follow-up bid, it will not cause the advertiser to bid too high and exceed the cost; nor will it cause the advertiser to bid too low and miss out on traffic.

[0162] To facilitate understanding of the solution, a specific embodiment is described below in conjunction with FIG. 3( a ) and FIG. 3( b ).

[0163] As shown in FIG3(a), on day T (any day), for the advertisement of the target APP with the second retention assessment requirement, if any user has visited it, the user belongs to the users in the activated population. At the same time, a visit record of the user can be formed, which indicates that the user has visited the advertisement of the target APP with the second retention assessment requirement.

[0164] On T+1 day, for the target user to be advertised, when advertising is being delivered, it is identified whether the target user to be advertised is a valid follow-up user (i.e., the above-mentioned ad follow-up user). If the identification result is yes, it indicates that the target user belongs to the follow-up traffic, otherwise, it indicates that the target user belongs to the non-follow-up traffic;

[0165] When the target user belongs to the follow-up traffic, for the same APP advertisement (i.e. the target advertisement about the target APP mentioned above), the first calculation formula is used to calculate the first type of evaluation index ECPM of the target advertisement 追投 For the remaining ads, i.e., other candidate ads, the second calculation formula is used to calculate the second evaluation index ECPM of the remaining ads 投放 ; Based on the calculation, the first type of evaluation index ECPM is obtained 追投 and the second type of evaluation indicator ECPM 投放, evaluate and rank the target advertisement and other candidate advertisements from high to low (i.e., ECPM ranking); based on the ranking results, select a specified number of advertisements with high rankings from the target advertisement and other candidate advertisements as the advertisements to be delivered; or, select evaluation indicators that exceed a predetermined threshold; randomly select a specified number of evaluation indicators from the evaluation indicators that exceed a predetermined threshold, and use the advertisements corresponding to the specified number of evaluation indicators as the advertisements to be delivered. That is, select the advertisement with the highest value as the advertisement to be delivered, and deliver the selected advertisement to be delivered to the target user (i.e., expose the advertisement with the highest ECPM). Among them, the first calculation formula is ECPM 追投 =price*pCTR*pCRR*(1-pDCVR), the second calculation formula is ECPM 投放 =price*pCTR*pCVR*pDCVR.

[0166] When the target user belongs to non-follow-up traffic, for each existing candidate advertisement, the second calculation formula is used to calculate the second evaluation index ECPM of each candidate advertisement 投放 , based on the calculated second evaluation index ECPM 投放 , sort the candidate ads from high to low (i.e., ECPM sorting), and based on the sorting results, select a specified number of ads with high rankings from the candidate ads as ads to be placed; or, select an evaluation index that exceeds a predetermined threshold, randomly select a specified number of evaluation indexes from the evaluation indexes that exceed the predetermined threshold, and select the ads corresponding to the specified number of evaluation indexes as ads to be placed. That is, select the ads with the highest values ​​as ads to be placed, and place the selected ads to be placed to the target users (i.e., expose the ads with the highest ECPM).

[0167] In addition, as shown in Figure 3(b), in the solution provided by the embodiment of the present invention, if the user activates the advertisement of the target APP with the second retention assessment requirement on the first day (the previous specified time period), then the user can join the activated group; on the second day, if it is necessary to deliver advertisements to the user, the advertising platform can perform intelligent follow-up delivery for the user and estimate the advertising evaluation indicators: estimate the evaluation indicators of the target advertisement (advertisement related to the target APP) as the follow-up advertisement, as well as the evaluation indicators of other candidate advertisements, and then, based on the calculated evaluation indicators, deliver targeted advertisements to the user.

[0168] In this scheme, when delivering advertisements to target users who are advertising chasing users, the first type of evaluation index is calculated for the target advertisement, and the second type of evaluation index is calculated for other candidate advertisements, and then the advertisements are evaluated and ranked based on the first type of evaluation index and the second type of evaluation index, so as to select the advertisements for delivery to the target users. Since the advertisements of the target APP can arouse the interest of the target users in the target APP, that is, there is a possibility of improving the secondary retention rate of the target APP, therefore, this scheme uses different types of evaluation indexes as considerations in the delivery process for different types of advertisements: advertisements that have an impact on the secondary retention rate and advertisements that do not have an impact, that is, advertisements related to the target APP and advertisements that are not related. It can be seen that through this scheme, for advertising chasing users, when delivering advertisements, different evaluation indicators are used as considerations in the delivery process for different types of advertisements. Therefore, the secondary retention rate of the APP of the network service customer can be improved while taking into account the user's stickiness to the advertising platform.

[0169] With respect to the above method embodiment, Figure 4 As shown, an embodiment of the present invention provides an advertisement delivery device, the device comprising:

[0170] The identification module 401 is used to identify whether the target user to be advertised is a follow-up advertising user; wherein the follow-up advertising user is a user who has visited the advertisement of the target APP in the last specified period; the target APP is the APP for which the retention rate is to be improved;

[0171] The calculation module 402 is used to calculate the first type of evaluation index of the target advertisement and the second type of evaluation index of other candidate advertisements other than the target advertisement if the recognition result is yes; wherein the target advertisement is the advertisement of the target APP, the first type of evaluation index is used to characterize the interest level of the target user in the advertisement re-delivered to the target APP, and the second type of evaluation index is used to characterize the interest level of the target user in other candidate advertisements other than the target advertisement;

[0172] A selection module 403 is used to select an advertisement to be placed from the target advertisement and other candidate advertisements based on the first evaluation index and the second evaluation index obtained by calculation;

[0173] The delivery module 404 is used to deliver the advertisement to the target user.

[0174] In this scheme, when delivering advertisements to target users who are advertising chasing users, the first type of evaluation index is calculated for the target advertisement, and the second type of evaluation index is calculated for other candidate advertisements, and then the advertisements are evaluated and ranked based on the first type of evaluation index and the second type of evaluation index, so as to select the advertisements for delivery to the target users. Since the advertisements of the target APP can arouse the interest of the target users in the target APP, that is, there is a possibility of improving the secondary retention rate of the target APP, therefore, this scheme uses different types of evaluation indexes as considerations in the delivery process for different types of advertisements: advertisements that have an impact on the secondary retention rate and advertisements that do not have an impact, that is, advertisements related to the target APP and advertisements that are not related. It can be seen that through this scheme, for advertising chasing users, when delivering advertisements, different evaluation indicators are used as considerations in the delivery process for different types of advertisements. Therefore, the secondary retention rate of the APP of the network service customer can be improved while taking into account the user's stickiness to the advertising platform.

[0175] Optionally, the target advertisement is: a historical advertisement of the target APP, or each advertisement of the target APP, each advertisement including all or part of the advertisement of the target APP;

[0176] The historical advertisements are advertisements visited by the target user during the last specified time period.

[0177] Optionally, the identification module is specifically used for:

[0178] Detecting whether there is a target history record of the target user, and if so, identifying the target user as an advertisement follow-up user;

[0179] The target history record is a record of visits to advertisements of a target APP with a retention assessment requirement within a previously specified period of time.

[0180] Optionally, the calculation module calculates a first type of evaluation index of the target advertisement, including:

[0181] Calculate a first type of evaluation index of the target advertisement based on a first estimated click-through rate, a click-to-buy rate, and a click-to-buy bid of the target advertisement;

[0182] Among them, the first estimated click-through rate is the probability that the target user clicks on the target advertisement after the target advertisement is released; the click-through rate is the probability that a successful follow-up occurs after the target user clicks on the target advertisement, and a successful follow-up is a click to enter the target APP; the follow-up bid is the bid required to be given by the advertiser of the target APP when the advertisement of the target APP is released again.

[0183] Optionally, the calculation module calculates a first type of evaluation index of the target advertisement based on a first estimated click-through rate, a click-to-buy rate, and a click-to-buy bid of the target advertisement, including:

[0184] Based on the first estimated click-through rate, click-to-buy rate, and buy-to-buy bid of the target advertisement, a first type of evaluation index of the target advertisement is calculated using a predetermined buy-to-buy formula;

[0185] Wherein, the predetermined follow-up investment formula includes:

[0186] ECPM 追投 =price 追投 *pCTR*pCRR;

[0187] Among them, ECPM 追投 For the first type of evaluation indicators, price 追投 is the follow-up bid, pCTR is the first estimated click-through rate, and pCRR is the click-through rate.

[0188] Optionally, the chasing bid is calculated by the following formula:

[0189] price 追投 =price*M;

[0190] Wherein, price is the bid of the target advertisement for the second retention conversion, M is the price correction factor and M=1-pDCVR, and pDCVR is the first deep conversion rate;

[0191] The first deep conversion rate is the probability that the target user completes the second retention after downloading the target APP.

[0192] Optionally, the click-through rate may be determined by:

[0193] Obtain the specified profile information of the target user and the specified attribute information of the target APP;

[0194] Calculate the click-through rate based on the specified portrait information and specified attribute information.

[0195] Optionally, the calculation module calculates the second type of evaluation indicators of other candidate advertisements except the target advertisement, including:

[0196] Calculate the second type of evaluation index of the other candidate advertisements based on the conversion bids, second estimated click-through rates, estimated conversion rates, and second deep conversion rates of the other candidate advertisements except the target advertisement;

[0197] The conversion bid is the bid of other candidate ads for achieving a specified conversion;

[0198] The second estimated click-through rate is the probability that the target user will click after other candidate ads are placed;

[0199] The estimated conversion rate is the probability that the target user will download the APP to which the other candidate ads belong after visiting the other candidate ads.

[0200] The second deep conversion rate is the probability that a specified conversion will occur after the target user downloads the APP to which other candidate ads belong.

[0201] Optionally, select a module specifically for:

[0202] Based on the calculated first-category evaluation index and second-category evaluation index, the target advertisement and other candidate advertisements are ranked from high to low in evaluation;

[0203] Based on the ranking results, a specified number of top-ranked advertisements are selected from the target advertisement and other candidate advertisements to obtain the advertisements to be delivered.

[0204] The embodiment of the present invention further provides an electronic device, such as Figure 5 As shown, it includes a processor 501, a communication interface 502, a memory 503 and a communication bus 504, wherein the processor 501, the communication interface 502, and the memory 503 communicate with each other through the communication bus 504.

[0205] Memory 503, used for storing computer programs;

[0206] The processor 501 is used to implement the steps of any one of the advertisement delivery methods when executing the program stored in the memory 503 .

[0207] The communication bus mentioned in the above terminal can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. The communication bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, only one thick line is used in the figure, but it does not mean that there is only one bus or one type of bus.

[0208] The communication interface is used for communication between the above terminal and other devices.

[0209] The memory may include a random access memory (RAM) or a non-volatile memory, such as at least one disk memory. Optionally, the memory may also be at least one storage device located away from the aforementioned processor.

[0210] The above-mentioned processor can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.

[0211] In another embodiment of the present invention, a computer-readable storage medium is provided, in which a computer program is stored. When the computer program is executed by a processor, the steps of any of the above-mentioned advertising delivery methods are implemented.

[0212] In another embodiment of the present invention, a computer program product including instructions is provided. When the computer program product is executed on a computer, the computer executes any one of the advertisement delivery methods in the above embodiments.

[0213] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented by software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function described in the embodiment of the present invention is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from a website site, computer, server or data center to another website site, computer, server or data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more available media integrated. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid-state hard disk Solid State Disk (SSD)), etc.

[0214] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

[0215] Each embodiment in this specification is described in a related manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the device and electronic device embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiments.

[0216] The above description is only a preferred embodiment of the present invention and is not intended to limit the protection scope of the present invention. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention are included in the protection scope of the present invention.

Claims

1. A method for placing advertisements, It is characterized in that include: Identify whether the target user to be advertised is a follow-up user; wherein the follow-up user is a user who has visited the advertisement of the target APP in the last specified period; the target APP is the APP for which the retention rate is to be improved; If the recognition result is yes, the first type of evaluation index of the target advertisement is calculated based on the first estimated click-through rate, click-to-follow-up rate and follow-up bid of the target advertisement, and the second type of evaluation index of the other candidate advertisements other than the target advertisement is calculated based on the conversion bid, second estimated click-through rate, estimated conversion rate and second deep conversion rate of other candidate advertisements other than the target advertisement; wherein, the target advertisement is the advertisement of the target APP, the first type of evaluation index is used to characterize the target user's interest in the advertisement that is re-delivered to the target APP, and the second type of evaluation index is used to characterize the target user's interest in the candidate advertisements other than the target advertisement; the first estimated click-through rate is the interest of the target user in the advertisement that is re-delivered to the target APP after the target advertisement is delivered. The probability of click; the click-through rate is the probability of a successful follow-up after the target user clicks on the target advertisement, and the successful follow-up means clicking into the target APP; the follow-up bid is the bid required by the advertiser of the target APP to re-run the advertisement of the target APP; the conversion bid is the bid of the other candidate advertisements for achieving a specified conversion; the second estimated click-through rate is the probability that the target user clicks after the other candidate advertisements are released; the estimated conversion rate is the probability that the target user downloads the APP to which the other candidate advertisements belong after visiting the other candidate advertisements; the second deep conversion rate is the probability that the target user can have the specified conversion after downloading the APP to which the other candidate advertisements belong; Selecting an advertisement to be placed from the target advertisement and other candidate advertisements based on the calculated first evaluation index and the second evaluation index; The advertisement to be delivered is delivered to the target user.

2. The method according to claim 1, It is characterized in that The target advertisement is a historical advertisement of the target APP, or each advertisement of the target APP, wherein each advertisement includes all or part of the advertisements of the target APP; The historical advertisements are advertisements visited by the target user during the last specified time period.

3. The method according to claim 1 or 2, It is characterized in that The step of identifying whether the target user to be placed an advertisement is a follow-up advertisement user includes: Detecting whether there is a target history record of the target user, and if so, identifying the target user as an advertisement follow-up user; The target history record is a record of visits to advertisements of a target APP with a retention assessment requirement within a previously specified period of time.

4. The method according to claim 1, It is characterized in that The calculating of the first evaluation index of the target advertisement based on the first estimated click-through rate, click-to-buy rate, and click-to-buy bid of the target advertisement includes: Based on the first estimated click-through rate, click-to-buy rate, and buy-to-buy bid of the target advertisement, a first type of evaluation index of the target advertisement is calculated using a predetermined buy-to-buy formula; Wherein, the predetermined follow-up investment formula includes: ECPM 追投 =price 追投 *pCTR*pCRR; Among them, ECPM 追投 For the first type of evaluation indicators, price 追投 is the follow-up bid, pCTR is the first estimated click-through rate, and pCRR is the click-through rate.

5. The method according to claim 4, It is characterized in that The follow-up bid is calculated by the following formula: price 追投 =price*M; Wherein, price is the bid of the target advertisement for the second retention conversion, M is the price correction factor and M=1-pDCVR, and pDCVR is the first deep conversion rate; The first deep conversion rate is the probability that the target user completes the second retention after downloading the target APP.

6. The method according to claim 1, It is characterized in that The click-through rate may be determined by: Obtaining the specified portrait information of the target user and the specified attribute information of the target APP; The click-through rate is calculated based on the designated portrait information and the designated attribute information.

7. The method according to claim 1 or 2, It is characterized in that Based on the first evaluation index and the second evaluation index obtained by calculation, selecting an advertisement to be delivered from the target advertisement and the other candidate advertisements includes: Based on the calculated first evaluation index and second evaluation index, the target advertisement and the other candidate advertisements are evaluated and ranked from high to low; Based on the ranking result, a specified number of advertisements with high rankings are selected from the target advertisement and the other candidate advertisements to obtain the advertisements to be delivered.

8. An advertising delivery device, It is characterized in that The device comprises: An identification module is used to identify whether the target user to be advertised is a follow-up advertising user; wherein the follow-up advertising user is a user who has visited the advertisement of the target APP within the last specified period; the target APP is the APP for which the retention rate is to be improved; A calculation module is used to calculate the first type of evaluation index of the target advertisement based on the first estimated click-through rate, click-to-follow-up rate and follow-up bid of the target advertisement if the recognition result is yes, and calculate the second type of evaluation index of the other candidate advertisements based on the conversion bid, second estimated click-through rate, estimated conversion rate and second deep conversion rate of other candidate advertisements other than the target advertisement; wherein the target advertisement is the advertisement of the target APP, the first type of evaluation index is used to characterize the interest level of the target user in the advertisement that is re-delivered to the target APP, and the second type of evaluation index is used to characterize the interest level of the target user in the candidate advertisements other than the target advertisement; the first estimated click-through rate is the interest level of the target user in the candidate advertisements other than the target advertisement after the target advertisement is delivered. the probability that the target user clicks on the target advertisement; the click-through rate is the probability that a successful follow-up occurs after the target user clicks on the target advertisement, and the successful follow-up means clicking into the target APP; the follow-up bid is the bid required by the advertiser of the target APP to re-run the advertisement of the target APP; the conversion bid is the bid of the other candidate advertisements for achieving a specified conversion; the second estimated click-through rate is the probability that the target user clicks after the other candidate advertisements are run; the estimated conversion rate is the probability that the target user downloads the APP to which the other candidate advertisements belong after visiting the other candidate advertisements; the second deep conversion rate is the probability that the target user can have the specified conversion after downloading the APP to which the other candidate advertisements belong; A selection module, configured to select an advertisement to be placed from the target advertisement and other candidate advertisements based on the first evaluation index and the second evaluation index obtained by calculation; A delivery module is used to deliver the advertisement to be delivered to the target user.

9. An electronic device, It is characterized in that It includes a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other through the communication bus; Memory, used to store computer programs; A processor, for implementing any of the methods described in claims 1-7 when executing a program stored in a memory.

10. A computer-readable storage medium, It is characterized in that The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.