Advertisement delivery method, device, electronic device and storage medium

Abstract

The present disclosure relates to an advertisement delivery method, device, electronic device and storage medium, the method comprising: when detecting that a target user account is a first type of user account, obtaining a target user feature of the target user account, the first type of user account being a user account with a registration duration less than a preset duration; determining a target retention rate of the target user account through the target user feature; when the target retention rate is less than the preset retention rate, adjusting the advertisement delivery density of the target user account, and delivering advertisements to the target user account through the adjusted advertisement delivery density, the preset retention rate being determined based on the retention rate of a second type of user account, the second type of user account being a user account with a registration duration greater than the preset duration. It can be seen that when the target retention rate is less than the preset retention rate, delivering advertisements to new user accounts through the adjusted advertisement delivery density can improve the retention rate of new user accounts on the platform and reduce the churn rate of new user accounts on the platform.

Description

Technical Field

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

[0002] With the continuous development of technology, many users can now watch or listen to multimedia programs provided by relevant platforms online through mobile devices such as mobile phones. For example, many users watch short videos online through short video platforms.

[0003] For platforms, it is usually necessary to continuously attract new users to increase the platform's user traffic. It usually takes some time for new users to register and become sticky to the platform. How to reduce the churn rate of new users during this period is an important operational indicator for the platform after attracting new users.

[0004] The inventors discovered in their research that the quantity and quality of advertising delivered to new users, from the time they register to when they become engaged with the platform, significantly influences whether they churn. Therefore, reducing the churn rate of new users through appropriate advertising is a pressing issue. Summary of the Invention

[0005] In order to reduce the churn rate of new users through appropriate advertising delivery, the embodiments of the present disclosure provide an advertising delivery method, device, electronic device, and storage medium.

[0006] According to a first aspect of an embodiment of the present disclosure, there is provided an advertisement delivery method, the method comprising:

[0007] When detecting that the target user account is a first type of user account, obtaining target user features of the target user account, the first type of user account being a user account with a registration duration less than a preset duration;

[0008] Determining a target retention rate of the target user account based on the target user characteristics;

[0009] When the target retention rate is lower than the preset retention rate, the advertising delivery density of the target user account is adjusted, and advertisements are delivered to the target user account using the adjusted advertising delivery density. The preset retention rate is determined based on the retention rate of the second type of user account, which is a user account with a registration duration greater than the preset duration.

[0010] Optionally, the target user characteristics are target user behavior characteristics corresponding to the target user account, and determining the target retention rate of the target user account based on the target user characteristics includes:

[0011] A behavior feature vector of the target user behavior feature is extracted, and a target retention rate of the target user account is determined based on a similarity between the extracted behavior feature vector and a historical behavior feature vector of historical user behavior features corresponding to historical first-category user accounts.

[0012] Optionally, the target user feature is a target user portrait feature corresponding to the target user account, and determining the target retention rate of the target user account based on the target user feature includes:

[0013] Extract the portrait feature vector of the target user portrait feature, and determine the target retention rate of the target user account based on the similarity between the extracted portrait feature vector and the historical portrait feature vector of the historical user portrait feature corresponding to the historical first type of user account.

[0014] Optionally, the target user characteristics are target user behavior characteristics and target user portrait characteristics corresponding to the target user account, and determining the target retention rate of the target user account based on the target user characteristics includes:

[0015] A target feature vector of the target user behavior feature and the target user portrait feature is extracted, and a target retention rate of the target user account is determined based on the similarity between the extracted feature vector and the historical feature vectors of the historical user behavior feature and the historical user portrait feature corresponding to the historical first category of user accounts.

[0016] Optionally, adjusting the advertisement delivery density of the target user account includes:

[0017] Get the initial advertising density of the target user account;

[0018] Determining an advertisement delivery density adjustment coefficient for the target user account based on the target retention rate, the preset retention rate, and the registration duration of the target user account;

[0019] The initial advertisement delivery density is adjusted using the advertisement delivery density adjustment coefficient to obtain an adjusted advertisement delivery density.

[0020] Optionally, when the difference between the target retention rate and the preset retention rate is the same, the longer the registration time of the target user account, the greater the advertising delivery density adjustment coefficient of the target user account;

[0021] When the registration duration of the target user accounts is the same, the greater the difference between the target retention rate and the preset retention rate, the smaller the advertising density adjustment coefficient of the target user accounts.

[0022] Optionally, determining the advertisement delivery density adjustment coefficient of the target user account according to the target retention rate, the preset retention rate, and the registration duration of the target user account includes:

[0023] The advertising density adjustment coefficient of the target user account is calculated according to the following formula:

[0024]

[0025] in, is the advertising adjustment coefficient, is the registration duration of the target user account, is the target retention rate, is the preset retention rate.

[0026] Optionally, the initial advertisement delivery density is obtained by:

[0027] Initial advertisement delivery density = the number of initially delivered advertisements / (the number of initially delivered advertisements + the number of initially delivered multimedia information).

[0028] According to a second aspect of an embodiment of the present disclosure, there is provided an advertisement delivery device, the device comprising:

[0029] A user feature acquisition module is configured to acquire target user features of the target user account when detecting that the target user account is a first type of user account, where the first type of user account is a user account whose registration time is less than a preset time;

[0030] a retention rate determination module, configured to determine a target retention rate of the target user account based on the target user characteristics;

[0031] The advertising delivery density adjustment module is configured to adjust the advertising delivery density of the target user account when the target retention rate is less than the preset retention rate, and deliver advertisements to the target user account through the adjusted advertising delivery density. The preset retention rate is determined based on the retention rate of the second type of user account, and the second type of user account is a user account with a registration duration greater than the preset duration.

[0032] Optionally, the target user feature is a target user behavior feature corresponding to the target user account, and the retention rate determination module is specifically configured to execute:

[0033] A behavior feature vector of the target user behavior feature is extracted, and a target retention rate of the target user account is determined based on a similarity between the extracted behavior feature vector and a historical behavior feature vector of historical user behavior features corresponding to historical first-category user accounts.

[0034] Optionally, the target user feature is a target user portrait feature corresponding to the target user account, and the retention rate determination module is specifically configured to execute:

[0035] Extract the portrait feature vector of the target user portrait feature, and determine the target retention rate of the target user account based on the similarity between the extracted portrait feature vector and the historical portrait feature vector of the historical user portrait feature corresponding to the historical first type of user account.

[0036] Optionally, the target user characteristics are target user behavior characteristics and target user portrait characteristics corresponding to the target user account, and the retention rate determination module is specifically configured to execute:

[0037] A target feature vector of the target user behavior feature and the target user portrait feature is extracted, and a target retention rate of the target user account is determined based on the similarity between the extracted feature vector and the historical feature vectors of the historical user behavior feature and the historical user portrait feature corresponding to the historical first category of user accounts.

[0038] Optionally, the advertisement delivery density adjustment module includes:

[0039] An initial advertisement delivery density obtaining unit configured to obtain an initial advertisement delivery density of a target user account;

[0040] an advertisement delivery density adjustment coefficient determining unit, configured to determine the advertisement delivery density adjustment coefficient of the target user account according to the target retention rate, the preset retention rate, and the registration duration of the target user account;

[0041] The adjusted advertisement delivery density obtaining unit is configured to adjust the initial advertisement delivery density using the advertisement delivery density adjustment coefficient to obtain the adjusted advertisement delivery density.

[0042] Optionally, when the difference between the target retention rate and the preset retention rate is the same, the longer the registration time of the target user account, the greater the advertising delivery density adjustment coefficient of the target user account;

[0043] When the registration duration of the target user accounts is the same, the greater the difference between the target retention rate and the preset retention rate, the smaller the advertising density adjustment coefficient of the target user accounts.

[0044] Optionally, the advertisement delivery density adjustment coefficient determination unit is specifically configured to execute:

[0045] The advertising density adjustment coefficient of the target user account is calculated according to the following formula:

[0046]

[0047] in, is the advertising adjustment coefficient, is the registration duration of the target user account, is the target retention rate, is the preset retention rate.

[0048] Optionally, the initial advertisement delivery density obtaining unit is configured to obtain the initial advertisement delivery density by:

[0049] Initial advertisement delivery density = the number of initially delivered advertisements / (the number of initially delivered advertisements + the number of initially delivered multimedia information).

[0050] According to a third aspect of an embodiment of the present disclosure, there is provided an electronic device, including:

[0051] processor;

[0052] a memory for storing instructions executable by the processor;

[0053] The processor is configured to execute the instructions to implement the advertisement delivery method described in the first aspect.

[0054] According to a fourth aspect of an embodiment of the present disclosure, a storage medium is provided. When instructions in the storage medium are executed by a processor of an electronic device, the electronic device can execute the advertisement delivery method described in the first aspect.

[0055] According to a fifth aspect of the embodiments of the present disclosure, a computer program product comprising instructions is provided, which, when executed on a computer, enables the computer to implement the advertisement delivery method described in the first aspect.

[0056] The technical solution provided by the embodiments of the present disclosure obtains target user characteristics of the target user account when it is detected that the target user account is a first-category user account, wherein the first-category user account is a user account with a registration duration less than a preset duration; determines a target retention rate of the target user account through the target user characteristics; when the target retention rate is less than the preset retention rate, adjusts the advertising delivery density of the target user account, and delivers advertisements to the target user account through the adjusted advertising delivery density, wherein the preset retention rate is determined based on the retention rate of the second-category user account, and the second-category user account is a user account with a registration duration greater than the preset duration.

[0057] It can be seen that through the technical solution provided by the embodiments of the present disclosure, when the target retention rate of a new user account is lower than the preset retention rate of an old user account, that is, when the new user account has not yet entered the mature stage on the platform and there is still a possibility of loss from the platform, advertisements can be delivered to the new user account through the adjusted advertising delivery density, thereby improving the retention rate of the new user account on the platform and reducing the loss rate of the new user account on the platform. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] Figure 1 is a flow chart showing an advertisement delivery method according to an exemplary embodiment;

[0059] Figure 2 yes Figure 1 A flowchart of an implementation of step S13 in the embodiment shown;

[0060] Figure 3 is a block diagram of an advertisement delivery device according to an exemplary embodiment;

[0061] Figure 4 is a block diagram of an electronic device according to an exemplary embodiment;

[0062] Figure 5 is a block diagram of an advertisement delivery device according to an exemplary embodiment;

[0063] Figure 6 The figure is a block diagram of another advertisement delivery device according to an exemplary embodiment. DETAILED DESCRIPTION

[0064] In order to enable ordinary persons in the art to better understand the technical solutions of the present disclosure, the technical solutions in the embodiments of the present disclosure will be clearly and completely described below with reference to the accompanying drawings.

[0065] It should be noted that the terms "first," "second," and the like in the specification and claims of the present disclosure and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or precedence. It should be understood that the numbers used in this manner are interchangeable where appropriate so that the embodiments of the present disclosure described herein can be implemented in an order other than those illustrated or described herein. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present disclosure. Instead, they are merely examples of apparatus and methods consistent with certain aspects of the present disclosure as detailed in the appended claims.

[0066] In order to reduce the churn rate of new users through appropriate advertising delivery, the embodiments of the present disclosure provide an advertising delivery method, device, electronic device, and storage medium.

[0067] In the first aspect, an advertisement delivery method provided by an embodiment of the present disclosure is first introduced in detail. The execution subject of the advertisement delivery method provided by an embodiment of the present disclosure can be an electronic device such as a server.

[0068] like Figure 1 As shown, an advertisement delivery method provided by an embodiment of the present disclosure may include the following steps:

[0069] In step S11 , when it is detected that the target user account is a first type of user account, a target user feature of the target user account is obtained.

[0070] Among them, the first type of user account is a user account whose registration time is less than the preset time.

[0071] Specifically, in actual applications, after receiving an access request from a target user account, it is possible to determine whether the target user account is a first-category user account, that is, to determine whether the target user account is a new user account whose registration time is less than a preset time. If it is detected that the target user account is a first-category user account, the target user characteristics of the target user account can be obtained.

[0072] The target user characteristics may include the user behavior characteristics corresponding to the target user account; or, the target user characteristics may also include the user portrait characteristics corresponding to the target user account; or, the target user characteristics may also include the user behavior characteristics and user portrait characteristics corresponding to the target user account. This is all reasonable, and the embodiments of the present disclosure do not specifically limit the target user characteristics.

[0073] Among them, the user behavior characteristics corresponding to the target user account may include: the longest daily usage time of the target user account since registration; the number of other user accounts currently followed by the target user account; the number of likes given by the target user account to works on the platform; the number of negative feedbacks given by the target user account to works on the platform, etc. The embodiments of the present disclosure do not specifically limit the user behavior characteristics corresponding to the target user account.

[0074] The user portrait features corresponding to the target user account may include: the registration duration of the target user account, the gender of the user corresponding to the target user account, the region corresponding to the target user account, etc. The embodiment of the present disclosure does not specifically limit the user portrait features corresponding to the target user account.

[0075] In step S12, the target retention rate of the target user account is determined based on the target user characteristics.

[0076] Since target user characteristics are used to characterize the characteristics of the target user account, once the target user characteristics of the target user account are obtained, the target user characteristics can be used to estimate the target user account's retention rate on the platform. For ease of description, the target user account's retention rate on the platform can be referred to as the target retention rate. It is understood that the target retention rate can be used to determine the stickiness between the current target user account and the platform. A higher target retention rate indicates a higher stickiness between the target user account and the platform; a lower target retention rate indicates a lower stickiness between the target user account and the platform.

[0077] In order to describe the solution clearly, the specific implementation method of determining the target retention rate of the target user account based on the target user characteristics will be introduced in detail in the following embodiments.

[0078] In step S13, when the target retention rate is less than the preset retention rate, the advertisement delivery density of the target user account is adjusted, and advertisements are delivered to the target user account using the adjusted advertisement delivery density.

[0079] Among them, the preset retention rate is determined based on the retention rate of the second type of user accounts, and the second type of user accounts are user accounts with a registration duration greater than the preset duration.

[0080] Specifically, the target retention rate can be used to characterize the current stickiness between the target user account and the platform. In order to further determine whether the current stickiness between the target user account and the platform reaches the stickiness between the old user account and the platform, it is necessary to compare the current stickiness between the target user account and the platform with the stickiness between the old user account and the platform. In other words, it is necessary to compare the target retention rate with the preset retention rate that can characterize the stickiness between the old user account and the platform.

[0081] Before comparing the target retention rate with the preset retention rate, the preset retention rate needs to be calculated. In one embodiment, the preset retention rate can be calculated by: calculating the retention rates of old user accounts within the past week, such as the 7-day retention rate; then calculating the weighted average of the retention rates of new user accounts within the past week; and determining the calculated weighted average as the preset retention rate. Of course, this is only one method of calculating the preset retention rate, and the embodiments of this disclosure do not specifically limit the method for calculating the preset retention rate.

[0082] After calculating the preset retention rate, the target retention rate can be compared with the preset retention rate. If the target retention rate is greater than or equal to the preset retention rate, it indicates that the target user account's current stickiness with the platform has reached the stickiness level of existing user accounts. The target user account has entered a mature stage on the platform. Adjusting the ad delivery density will have little effect on improving the target user account's retention rate. Therefore, the ad delivery density for the target user account can be left unchanged.

[0083] If the target retention rate is lower than the preset retention rate, it means that the current stickiness of the target user account with the platform has not yet reached the stickiness level of the old user account, and the target user account has not yet entered the mature stage on the platform. Adjusting the advertising density will have a higher benefit in improving the retention rate of the target user account. Therefore, the advertising density of the target user account can be adjusted, and advertisements can be delivered to the target user account through the adjusted advertising density, thereby improving the retention rate of the target user account and reducing the churn rate of the target user account on the platform.

[0084] The advertisement delivery density can be the number of delivered advertisements divided by the sum of the number of delivered advertisements and the number of multimedia messages, i.e., advertisement delivery density = number of delivered advertisements / (number of delivered advertisements + number of multimedia messages). The multimedia messages can be videos, music, image collections, etc., and the embodiments of the present disclosure do not specifically limit the multimedia messages.

[0085] For a clearer description of the solution, the specific implementation method of adjusting the advertisement delivery density of the target user account will be described in detail in the following embodiments.

[0086] The technical solution provided by the embodiments of the present disclosure obtains target user characteristics of the target user account when it is detected that the target user account is a first-category user account, wherein the first-category user account is a user account with a registration duration less than a preset duration; determines a target retention rate of the target user account through the target user characteristics; when the target retention rate is less than the preset retention rate, adjusts the advertising delivery density of the target user account, and delivers advertisements to the target user account through the adjusted advertising delivery density, wherein the preset retention rate is determined based on the retention rate of the second-category user account, and the second-category user account is a user account with a registration duration greater than the preset duration.

[0087] It can be seen that through the technical solution provided by the embodiments of the present disclosure, when the target retention rate of a new user account is lower than the preset retention rate of an old user account, that is, when the new user account has not yet entered the mature stage on the platform and there is still a possibility of loss from the platform, advertisements can be delivered to the new user account through the adjusted advertising delivery density, thereby improving the retention rate of the new user account on the platform and reducing the loss rate of the new user account on the platform.

[0088] For a clearer description of the solution, determining the target retention rate of a target user account by using target user characteristics will be described in detail in the following embodiments.

[0089] In one embodiment, the target user characteristics are target user behavior characteristics corresponding to the target user account. In this case, determining the target retention rate of the target user account based on the target user characteristics may include the following steps:

[0090] A behavior feature vector of the target user behavior feature is extracted, and a target retention rate of the target user account is determined based on the similarity between the extracted behavior feature vector and the historical behavior feature vector of the historical user behavior feature corresponding to the historical first category of user accounts.

[0091] In this embodiment, the target user characteristics are target user behavior characteristics corresponding to the target user account. It should be noted that the user behavior characteristics have been described in detail in the above step S11, and will not be repeated here.

[0092] It can be understood that the historical behavior feature vector and retention rate of the historical user behavior characteristics of the historical first-category user account, i.e., the historical new user account, are known. Therefore, the target retention rate of the target user account can be determined by the similarity between the behavior feature vector of the target user behavior characteristics and the historical behavior feature vector.

[0093] As an implementation method of the embodiment of the present disclosure, a behavioral feature vector of the target user's behavioral features can be extracted, and the similarity between the behavioral feature vector of the target user's behavioral features and the historical behavioral feature vector of the historical user behavioral features of the historical first-category user account can be calculated; if it is calculated that the similarity between the behavioral feature vector of the target user's behavioral features and the historical behavioral feature vector of the historical user behavioral features of a certain historical first-category user account is high, then the retention rate of the historical first-category user account can be determined as the target retention rate of the target user account.

[0094] As another implementation method of the embodiment of the present disclosure, in order to accurately determine the target retention rate of the target user account, the historical user behavior characteristics corresponding to the historical first-category user accounts and the retention rates of the historical first-category user accounts can be used in advance to train the retention rate prediction model.

[0095] Specifically, since the historical user behavior characteristics and retention rates of historical first-category user accounts are both known, a large number of user behavior characteristics and retention rates of historical first-category user accounts can be collected. The historical user behavior characteristics of historical first-category user accounts are used as input to a preset model, and the retention rates of historical first-category user accounts are used as output of the preset model to train the preset model. During the training of the preset model, the preset model extracts the historical behavior feature vectors of the historical user behavior characteristics and continuously adjusts the model parameters of the preset model. When the accuracy of the output retention rate of the historical first-category user accounts reaches the preset accuracy, the model parameters of the preset model are determined, thereby obtaining a trained retention rate prediction model.

[0096] After obtaining the trained retention rate prediction model, the target user behavior characteristics can be input into the pre-trained retention rate prediction model. The pre-trained retention rate preset model extracts the behavior feature vector of the target user behavior characteristics and uses algorithms such as logistic regression used when training the retention rate preset model to calculate the target retention rate corresponding to the behavior feature vector, that is, output the target retention rate from the pre-trained retention rate prediction model.

[0097] It can be understood that the pre-trained retention rate prediction model is trained based on the historical behavior feature vector of historical user behavior characteristics and the retention rate of historical first-category user accounts. Therefore, the higher the similarity between the behavior feature vector of the target user behavior characteristics and the historical behavior feature vector of the historical user behavior characteristics corresponding to a first-category user account, the closer the target retention rate output from the pre-trained retention rate prediction model is to the retention rate of the first-category user account; otherwise, the target retention rate output from the trained retention rate prediction model is more different from the retention rate of the first-category user account.

[0098] It should be noted that the retention rate of historical first-category user accounts can be: the percentage of historical first-category user accounts that were active on the platform on day N+7, of the set of historical first-category user accounts that were active on the platform. N can be determined based on specific circumstances and is not specifically limited in this embodiment. Furthermore, the preset model can be determined based on actual circumstances and is not specifically limited in this embodiment.

[0099] As can be seen from the above description, the retention rate prediction model can predict the retention rate of new user accounts. Therefore, after inputting the target user behavior characteristics corresponding to the target user account into the retention rate prediction model, the target retention rate of the target user account can be accurately obtained. This target retention rate can be the probability that the target user account will still remain on the platform after 7 days.

[0100] In another embodiment, the target user characteristics are target user profile characteristics corresponding to the target user account. In this case, determining the target retention rate of the target user account based on the target user characteristics may include the following steps:

[0101] A portrait feature vector of the target user portrait feature is extracted, and a target retention rate of the target user account is determined based on the similarity between the extracted portrait feature vector and the historical portrait feature vector of the historical user portrait feature corresponding to the historical first category of user accounts.

[0102] In this embodiment, the target user feature is the target user portrait feature corresponding to the target user account. It should be noted that the user portrait feature has been described in detail in the above step S11, and will not be repeated here.

[0103] It can be understood that the historical portrait feature vector and retention rate of the historical user portrait features of the historical first category of user accounts, i.e., historical new user accounts, are known. Therefore, the target retention rate of the target user account can be determined by the similarity between the portrait feature vector of the target user portrait feature and the historical portrait feature vector.

[0104] As an implementation method of the embodiment of the present disclosure, a portrait feature vector of the target user portrait feature can be extracted, and the similarity between the portrait feature vector of the target user portrait feature and the historical portrait feature vector of the historical user portrait feature of the historical first-category user account can be calculated; if the calculated similarity between the portrait feature vector of the target user portrait feature and the historical portrait feature vector of the historical user portrait feature of a certain historical first-category user account is high, then the retention rate of the historical first-category user account can be determined as the target retention rate of the target user account.

[0105] As another implementation method of the embodiment of the present disclosure, in order to accurately determine the target retention rate of the target user account, the historical user portrait features corresponding to the historical first-category user accounts and the retention rates of the historical first-category user accounts can be used in advance to train the retention rate prediction model.

[0106] Specifically, since the historical user profile features and retention rates of historical first-category user accounts are both known, a large number of user profile features and retention rates of historical first-category user accounts can be collected. The historical user profile features of historical first-category user accounts are used as input to a preset model, and the retention rates of historical first-category user accounts are used as output of the preset model to train the preset model. During the training of the preset model, the preset model extracts the historical profile feature vectors of the historical user profile features and continuously adjusts the model parameters of the preset model. When the accuracy of the output retention rate of the historical first-category user accounts reaches the preset accuracy, the model parameters of the preset model are determined, thereby obtaining a trained retention rate prediction model.

[0107] After obtaining the trained retention rate prediction model, the target user portrait features can be input into the pre-trained retention rate prediction model. The pre-trained retention rate preset model extracts the portrait feature vector of the target user portrait features and uses algorithms such as logistic regression used when training the retention rate preset model to calculate the target retention rate corresponding to the portrait feature vector, that is, output the target retention rate from the pre-trained retention rate prediction model.

[0108] It can be understood that the pre-trained retention rate prediction model is trained based on the historical portrait feature vector of historical user portrait features and the retention rate of historical first-category user accounts. Therefore, the higher the similarity between the portrait feature vector of the target user portrait feature and the historical portrait feature vector of the historical user portrait feature corresponding to a first-category user account, the closer the target retention rate output from the trained retention rate prediction model is to the retention rate of the first-category user account; otherwise, the target retention rate output from the trained retention rate prediction model is more different from the retention rate of the first-category user account.

[0109] It should be noted that the retention rate of historical first-category user accounts can be: the percentage of historical first-category user accounts that were active on the platform on day N+7, of the set of historical first-category user accounts that were active on the platform. N can be determined based on specific circumstances and is not specifically limited in this embodiment. Furthermore, the preset model can be determined based on actual circumstances and is not specifically limited in this embodiment.

[0110] As can be seen from the above description, the retention rate prediction model can predict the retention rate of new user accounts. Therefore, after inputting the target user profile features corresponding to the target user account into the retention rate prediction model, the target retention rate of the target user account can be accurately obtained. The target retention rate can be the probability that the target user account will still remain on the platform after 7 days.

[0111] In another embodiment, the target user characteristics are target user behavior characteristics and target user profile characteristics corresponding to the target user account. In this case, determining the target retention rate of the target user account based on the target user characteristics may include the following steps:

[0112] A target feature vector of the target user behavior features and the target user portrait features is extracted, and a target retention rate of the target user account is determined based on the similarity between the extracted target feature vector and the historical feature vectors of the historical user behavior features and the historical user portrait features corresponding to the historical first category of user accounts.

[0113] In this embodiment, the target user features are the historical user behavior features and target user profile features corresponding to the target user account. It can be seen that in this embodiment, the dimensionality of the target user features has increased. Therefore, compared with the behavior feature vector or profile feature vector described in the above embodiment, the dimensionality of the target feature vector has also increased, which is conducive to obtaining a more accurate target retention rate. It should be noted that the user behavior features and user profile features have been explained in detail in step S11 above, and will not be repeated here.

[0114] It is understandable that the historical feature vectors and retention rates of the historical first-category user accounts, i.e., historical new user accounts, are known. Therefore, the target retention rate of the target user account can be determined by the similarity between the target feature vector and the historical feature vector.

[0115] As an implementation method of an embodiment of the present disclosure, a target feature vector can be extracted, and the similarity between the target feature vector and the historical feature vector can be calculated; if the calculated similarity between the target feature vector and the historical feature vector of a historical first-category user account is high, then the retention rate of the historical first-category user account can be determined as the target retention rate of the target user account.

[0116] As another implementation method of the embodiment of the present disclosure, in order to accurately determine the target retention rate of the target user account, the historical user behavior characteristics and historical user portrait characteristics corresponding to the historical first-category user accounts, as well as the retention rate of the historical first-category user accounts, can be used in advance to train the retention rate prediction model.

[0117] Specifically, since the historical user behavior characteristics and historical user portrait characteristics of historical first-category user accounts, as well as the retention rates of historical first-category user accounts, are all known, it is possible to collect a large number of historical user behavior characteristics and user portrait characteristics of historical first-category user accounts, as well as the retention rates of historical first-category user accounts. The historical user behavior characteristics and historical user portrait characteristics of historical first-category user accounts are used as inputs to a preset model, and the retention rates of historical first-category user accounts are used as outputs of the preset model to train the preset model. During the process of training the preset model, the preset model extracts historical feature vectors of the historical user behavior characteristics and historical user portrait characteristics, and continuously adjusts the model parameters of the preset model. When the accuracy of the output retention rate of the historical first-category user accounts reaches the preset accuracy, the model parameters of the preset model are determined to obtain a trained retention rate prediction model.

[0118] After obtaining the trained retention rate prediction model, the historical user behavior characteristics and the target user portrait characteristics can be input into the pre-trained retention rate prediction model. The pre-trained retention rate preset model extracts the target feature vector and uses the logistic regression and other algorithms used when training the retention rate preset model to calculate the target retention rate corresponding to the target feature vector, that is, output the target retention rate from the pre-trained retention rate prediction model.

[0119] It can be understood that the pre-trained retention rate prediction model is trained based on historical feature vectors and the retention rates of historical first-category user accounts. Therefore, the higher the similarity between the target feature vector and the historical feature vector of a first-category user account, the closer the target retention rate output from the pre-trained retention rate prediction model is to the retention rate of the first-category user account; otherwise, the target retention rate output from the trained retention rate prediction model will differ more greatly from the retention rate of the first-category user account.

[0120] It should be noted that the retention rate of historical first-category user accounts can be: the percentage of historical first-category user accounts that were active on the platform on day N+7, of the set of historical first-category user accounts that were active on the platform. N can be determined based on specific circumstances and is not specifically limited in this embodiment. Furthermore, the preset model can be determined based on actual circumstances and is not specifically limited in this embodiment.

[0121] As can be seen from the above description, the retention rate prediction model can predict the retention rate of new user accounts. Therefore, after inputting the target user behavior characteristics and target user profile characteristics corresponding to the target user account into the retention rate prediction model, the target retention rate of the target user account can be accurately obtained. This target retention rate can be regarded as the probability that the target user account will still remain on the platform after 7 days.

[0122] In order to describe the solution clearly, the specific implementation method of adjusting the advertisement delivery density of the target user account will be introduced in detail below.

[0123] In one embodiment, the advertisement delivery density of the target user account is adjusted, such as Figure 2 As shown, the following steps may be included:

[0124] In step S21, the initial advertisement delivery density of the target user account is obtained.

[0125] Specifically, the initial advertisement delivery density may be determined based on the number of advertisements and multimedia information initially delivered to the target user account after the target user account is registered.

[0126] In one embodiment, the initial advertisement delivery density can be obtained as follows: initial advertisement delivery density = number of initially delivered advertisements / (number of initially delivered advertisements + number of initially delivered multimedia messages). Thus, the initial advertisement delivery density can be accurately calculated using the number of initially delivered advertisements and the number of initially delivered multimedia messages.

[0127] In step S22, the advertisement delivery density adjustment coefficient of the target user account is determined according to the target retention rate, the preset retention rate, and the registration duration of the target user account.

[0128] The ad density adjustment coefficient is typically between 0 and 1. A larger ad density adjustment coefficient indicates a smaller adjustment to the initial ad density. Furthermore, when the ad density adjustment coefficient is 1, no adjustment to the initial ad density is made. A smaller ad density adjustment coefficient indicates a larger adjustment to the initial ad density.

[0129] Moreover, the advertising density adjustment coefficient is not only related to the difference between the target retention rate and the preset retention rate, but also to the registration duration of the target user account.

[0130] Specifically, the smaller the difference between the target retention rate and the preset retention rate, the higher the stickiness between the target user account and the platform, and the more familiar the target user account is with the platform. The benefit of improving the retention rate of the target user account by adjusting the advertising density is usually lower. Therefore, the advertising density adjustment coefficient of the target user account can be larger, that is, it can be close to 1, and the adjustment of the initial advertising density in subsequent steps will be smaller.

[0131] The greater the difference between the target retention rate and the preset retention rate, the lower the stickiness between the target user account and the platform, and the less familiar the target user account is with the platform. Adjusting the advertising density to improve the retention rate of the target user account is usually more profitable. Therefore, the advertising density adjustment coefficient of the target user account can be smaller, and the adjustment of the initial advertising density in subsequent steps can be larger.

[0132] The longer the target user account is registered, the more familiar the target user account is with the platform. The benefits of improving the target user account's retention rate by adjusting the advertising density are usually lower. Therefore, the advertising density adjustment coefficient of the target user account can be larger, that is, it can be close to 1, and the adjustment of the initial advertising density in subsequent steps will be smaller.

[0133] The shorter the registration time of the target user account, the less familiar the target user account is with the platform. Adjusting the advertising density will usually have a higher benefit in improving the retention rate of the target user account. Therefore, the advertising density adjustment coefficient of the target user account can be smaller, and the adjustment of the initial advertising density in subsequent steps can be larger.

[0134] From the above analysis, it can be seen that as an implementation method of the embodiment of the present disclosure, when the difference between the target retention rate and the preset retention rate is the same, the longer the registration time of the target user account, the greater the advertising density adjustment coefficient of the target user account.

[0135] When the registration duration of the target user accounts is the same, the greater the difference between the target retention rate and the preset retention rate, the smaller the advertising density adjustment coefficient of the target user accounts.

[0136] In this embodiment, when the difference between the target retention rate and the preset retention rate is the same, it means that the target user accounts currently have the same stickiness to the platform, and the target user accounts with longer registration time have higher maturity on the platform. The benefit of adjusting the advertising density to improve the retention rate of target user accounts with longer registration time is usually lower. Therefore, the larger the advertising density adjustment coefficient of the target user accounts with longer registration time, the smaller the adjustment of the initial advertising density in subsequent steps.

[0137] When the registration duration of the target user accounts is the same, the greater the difference between the target retention rate and the preset retention rate, the lower the current stickiness of the target user account to the platform. By adjusting the advertising density, the benefits of adjusting the retention rate of the target user account with a larger difference between the target retention rate and the preset retention rate are usually higher. Therefore, the smaller the advertising density adjustment coefficient of the target user account, the greater the adjustment of the initial advertising density in the subsequent steps.

[0138] As an implementation of the embodiment of the present disclosure, determining an advertisement delivery density adjustment coefficient for a target user account based on a target retention rate, a preset retention rate, and the registration duration of the target user account may include the following steps:

[0139] Calculate the ad delivery density adjustment coefficient for the target user account using the following formula:

[0140]

[0141] in, is the advertising adjustment coefficient, is the registration duration of the target user account, is the target retention rate, Of course, this calculation formula is only used as an example to illustrate the calculation method of the advertisement delivery density adjustment coefficient, and the embodiment of the present disclosure does not make any specific limitation on this.

[0142] In step S23, the initial advertisement delivery density is adjusted using the advertisement delivery density adjustment coefficient to obtain an adjusted advertisement delivery density.

[0143] From the above description, we can see that if the target user account has a higher stickiness to the platform, then the ad density adjustment coefficient will be larger, and the adjusted ad density will be approximately the same as the initial ad density, that is, the adjustment to the initial ad density will be smaller. If the target user account has a lower stickiness to the platform, then the ad density adjustment coefficient will be smaller, and the difference between the adjusted ad density and the initial ad density will be larger, that is, the adjustment to the initial ad density will be larger. Therefore, by adjusting the ad density of target user accounts with lower platform stickiness, the retention rate of target user accounts with lower platform stickiness can be improved, that is, the churn rate of target user accounts with lower platform stickiness can be reduced.

[0144] The adjusted advertising delivery density may be: advertising delivery density adjustment coefficient * initial advertising delivery density.

[0145] Specifically, since the size of the advertising delivery density adjustment coefficient is usually between 0 and 1, the adjusted advertising delivery density is usually lower than the initial advertising delivery density, thereby reducing the loss of target user accounts with low platform stickiness due to excessive advertising delivery to target user accounts with low platform stickiness, thereby improving the retention rate of target user accounts with low platform stickiness.

[0146] According to a second aspect of the embodiments of the present disclosure, an advertisement delivery device is provided. Figure 3 As shown, the device includes:

[0147] The user feature acquisition module 310 is configured to acquire target user features of the target user account when detecting that the target user account is a first type of user account, where the first type of user account is a user account whose registration time is less than a preset time;

[0148] A retention rate determination module 320 is configured to determine a target retention rate of the target user account based on the target user characteristics;

[0149] The advertising delivery density adjustment module 330 is configured to adjust the advertising delivery density of the target user account when the target retention rate is less than the preset retention rate, and deliver advertisements to the target user account through the adjusted advertising delivery density. The preset retention rate is determined based on the retention rate of the second type of user account, and the second type of user account is a user account with a registration duration greater than the preset duration.

[0150] It can be seen that through the technical solution provided by the embodiments of the present disclosure, when the target retention rate of a new user account is lower than the preset retention rate of an old user account, that is, when the new user account has not yet entered the mature stage on the platform and there is still a possibility of loss from the platform, advertisements can be delivered to the new user account through the adjusted advertising delivery density, thereby improving the retention rate of the new user account on the platform and reducing the loss rate of the new user account on the platform.

[0151] Optionally, the target user feature is a target user behavior feature corresponding to the target user account, and the retention rate determination module is specifically configured to execute:

[0152] A behavior feature vector of the target user behavior feature is extracted, and a target retention rate of the target user account is determined based on a similarity between the extracted behavior feature vector and a historical behavior feature vector of historical user behavior features corresponding to historical first-category user accounts.

[0153] Optionally, the target user feature is a target user portrait feature corresponding to the target user account, and the retention rate determination module is specifically configured to execute:

[0154] Extract the portrait feature vector of the target user portrait feature, and determine the target retention rate of the target user account based on the similarity between the extracted portrait feature vector and the historical portrait feature vector of the historical user portrait feature corresponding to the historical first type of user account.

[0155] Optionally, the target user characteristics are target user behavior characteristics and target user portrait characteristics corresponding to the target user account, and the retention rate determination module is specifically configured to execute:

[0156] A target feature vector of the target user behavior feature and the target user portrait feature is extracted, and a target retention rate of the target user account is determined based on the similarity between the extracted feature vector and the historical feature vectors of the historical user behavior feature and the historical user portrait feature corresponding to the historical first category of user accounts.

[0157] Optionally, the advertisement delivery density adjustment module includes:

[0158] An initial advertisement delivery density obtaining unit configured to obtain an initial advertisement delivery density of a target user account;

[0159] an advertisement delivery density adjustment coefficient determining unit, configured to determine the advertisement delivery density adjustment coefficient of the target user account according to the target retention rate, the preset retention rate, and the registration duration of the target user account;

[0160] The adjusted advertisement delivery density obtaining unit is configured to adjust the initial advertisement delivery density using the advertisement delivery density adjustment coefficient to obtain the adjusted advertisement delivery density.

[0161] Optionally, when the difference between the target retention rate and the preset retention rate is the same, the longer the registration time of the target user account, the greater the advertising delivery density adjustment coefficient of the target user account;

[0162] When the registration duration of the target user accounts is the same, the greater the difference between the target retention rate and the preset retention rate, the smaller the advertising density adjustment coefficient of the target user accounts.

[0163] Optionally, the advertisement delivery density adjustment coefficient determination unit is specifically configured to execute:

[0164] The advertising density adjustment coefficient of the target user account is calculated according to the following formula:

[0165]

[0166] in, is the advertising adjustment coefficient, is the registration duration of the target user account, is the target retention rate, is the preset retention rate.

[0167] Optionally, the initial advertisement delivery density obtaining unit is configured to obtain the initial advertisement delivery density by:

[0168] Initial advertisement delivery density = the number of initially delivered advertisements / (the number of initially delivered advertisements + the number of initially delivered multimedia information).

[0169] According to a third aspect of the embodiments of the present disclosure, there is provided an electronic device, such as Figure 4 Shown, including:

[0170] Processor 410;

[0171] Memory 420 for storing the processor executable instructions;

[0172] The processor is configured to execute the instructions to implement the advertisement delivery method described in the first aspect.

[0173] It can be seen that through the technical solution provided by the embodiments of the present disclosure, when the target retention rate of a new user account is lower than the preset retention rate of an old user account, that is, when the new user account has not yet entered the mature stage on the platform and there is still a possibility of loss from the platform, advertisements can be delivered to the new user account through the adjusted advertising delivery density, thereby improving the retention rate of the new user account on the platform and reducing the loss rate of the new user account on the platform.

[0174] Figure 5 FIG. 5 is a block diagram of an advertisement delivery device 500 according to an exemplary embodiment. For example, the device 500 may be provided as a server. Figure 5 Apparatus 500 includes a processing component 522, which further includes one or more processors, and memory resources represented by memory 532 for storing instructions executable by processing component 522, such as applications. The applications stored in memory 532 may include one or more modules, each corresponding to a set of instructions. Furthermore, processing component 522 is configured to execute the instructions to perform the advertising delivery method described in the first aspect.

[0175] The device 500 may also include a power supply component 526 configured to perform power management of the device 500, a wired or wireless network interface 550 configured to connect the device 500 to a network, and an input / output (I / O) interface 558. The device 500 may operate based on an operating system stored in the memory 532, such as Windows Server™, Mac OS X™, Unix™, Linux™, FreeBSD™, or the like.

[0176] It can be seen that through the technical solution provided by the embodiments of the present disclosure, when the target retention rate of a new user account is lower than the preset retention rate of an old user account, that is, when the new user account has not yet entered the mature stage on the platform and there is still a possibility of loss from the platform, advertisements can be delivered to the new user account through the adjusted advertising delivery density, thereby improving the retention rate of the new user account on the platform and reducing the loss rate of the new user account on the platform.

[0177] Figure 6 FIG6 is a block diagram of an apparatus 600 for delivering advertisements according to an exemplary embodiment. For example, the apparatus 600 may be a mobile phone, a computer, a digital broadcast electronic device, a messaging device, a game console, a tablet device, a medical device, a fitness device, a personal digital assistant, etc.

[0178] Reference Figure 6 , apparatus 600 may include one or more of the following components: a processing component 602 , a memory 604 , a power component 606 , a multimedia component 608 , an audio component 610 , an input / output (I / O) interface 612 , a sensor component 614 , and a communication component 616 .

[0179] The processing component 602 generally controls the overall operation of the device 600, such as operations associated with display, phone calls, data communications, camera operation, and recording operations. The processing component 602 may include one or more processors 620 to execute instructions to perform all or part of the steps of the above-described method. In addition, the processing component 602 may include one or more modules to facilitate interaction between the processing component 602 and other components. For example, the processing component 602 may include a multimedia module to facilitate interaction between the multimedia component 608 and the processing component 602.

[0180] The memory 604 is configured to store various types of data to support operations on the device 600. Examples of such data include instructions for any application or method operating on the device 600, contact data, phone book data, messages, pictures, videos, etc. The memory 604 can be implemented by any type of volatile or non-volatile storage device, or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk, or optical disk.

[0181] The power supply component 607 provides power to the various components of the device 600. The power supply component 607 may include a power management system, one or more power supplies, and other components associated with generating, managing, and distributing power to the device 600.

[0182] The multimedia component 608 includes a screen that provides an output interface between the device 600 and the user. In some embodiments, the screen may include a liquid crystal display (LCD) and a touch panel (TP). If the screen includes a touch panel, the screen may be implemented as a touch screen to receive input signals from the user. The touch panel includes one or more touch sensors to sense touches, slides, and gestures on the touch panel. The touch sensors can not only sense the boundaries of a touch or slide action, but also detect the duration and pressure associated with the touch or slide action. In some embodiments, the multimedia component 608 includes a front-facing camera and / or a rear-facing camera. When the device 600 is in an operating mode, such as a capture mode or a video mode, the front-facing camera and / or the rear-facing camera can receive external multimedia data. Each front-facing camera and the rear-facing camera can have a fixed optical lens system or have focal length and optical zoom capabilities.

[0183] The audio component 610 is configured to output and / or input audio signals. For example, the audio component 610 includes a microphone (MIC) that is configured to receive external audio signals when the device 600 is in an operating mode, such as a call mode, a recording mode, or a voice recognition mode. The received audio signals may be further stored in the memory 404 or transmitted via the communication component 616. In some embodiments, the audio component 610 also includes a speaker for outputting audio signals.

[0184] I / O interface 612 provides an interface between processing component 602 and peripheral interface modules, such as a keyboard, click wheel, buttons, etc. These buttons may include but are not limited to: a home button, volume buttons, a start button, and a lock button.

[0185] The sensor assembly 614 includes one or more sensors for providing various aspects of the status assessment of the device 600. For example, the sensor assembly 614 can detect the open / closed state of the device 600, the relative positioning of components, such as the display and keypad of the device 600. The sensor assembly 614 can also detect changes in the position of the device 600 or a component of the device 600, the presence or absence of user contact with the device 600, the orientation or acceleration / deceleration of the device 600, and temperature changes of the device 600. The sensor assembly 614 may include a proximity sensor configured to detect the presence of nearby objects without any physical contact. The sensor assembly 614 may also include an optical sensor, such as a CMOS or CCD image sensor, for use in imaging applications. In some embodiments, the sensor assembly 614 may also include an accelerometer, a gyroscope, a magnetic sensor, a pressure sensor, or a temperature sensor.

[0186] The communication component 616 is configured to facilitate wired or wireless communication between the apparatus 600 and other devices. The apparatus 600 can access a wireless network based on a communication standard, such as WiFi, a carrier network (such as 2G, 3G, 4G or 5G), or a combination thereof. In an exemplary embodiment, the communication component 416 receives a broadcast signal or broadcast-related information from an external broadcast management system via a broadcast channel. In an exemplary embodiment, the communication component 616 also includes a near field communication (NFC) module to facilitate short-range communication. For example, the NFC module can be implemented based on radio frequency identification (RFID) technology, infrared data association (IrDA) technology, ultra-wideband (UWB) technology, Bluetooth (BT) technology, and other technologies.

[0187] In an exemplary embodiment, the device 600 can be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors or other electronic components to execute the advertising delivery method described in the first aspect.

[0188] In an exemplary embodiment, a non-transitory computer-readable storage medium including instructions is also provided, such as a memory 604 including instructions. The instructions can be executed by the processor 620 of the apparatus 600 to perform the above method. Alternatively, for example, the storage medium can be a non-transitory computer-readable storage medium, such as a ROM, a random access memory (RAM), a CD-ROM, a magnetic tape, a floppy disk, an optical data storage device, and the like.

[0189] It can be seen that through the technical solution provided by the embodiments of the present disclosure, when the target retention rate of a new user account is lower than the preset retention rate of an old user account, that is, when the new user account has not yet entered the mature stage on the platform and there is still a possibility of loss from the platform, advertisements can be delivered to the new user account through the adjusted advertising delivery density, thereby improving the retention rate of the new user account on the platform and reducing the loss rate of the new user account on the platform.

[0190] According to a fourth aspect of an embodiment of the present disclosure, a storage medium is provided. When instructions in the storage medium are executed by a processor of an electronic device, the electronic device can execute the advertisement delivery method described in the first aspect.

[0191] It can be seen that through the technical solution provided by the embodiments of the present disclosure, when the target retention rate of a new user account is lower than the preset retention rate of an old user account, that is, when the new user account has not yet entered the mature stage on the platform and there is still a possibility of loss from the platform, advertisements can be delivered to the new user account through the adjusted advertising delivery density, thereby improving the retention rate of the new user account on the platform and reducing the loss rate of the new user account on the platform.

[0192] According to a tenth aspect of the embodiments of the present disclosure, a computer program product comprising instructions is provided, which, when executed on a computer, enables the computer to implement the advertisement delivery method described in the first aspect.

[0193] It can be seen that through the technical solution provided by the embodiments of the present disclosure, when the target retention rate of a new user account is lower than the preset retention rate of an old user account, that is, when the new user account has not yet entered the mature stage on the platform and there is still a possibility of loss from the platform, advertisements can be delivered to the new user account through the adjusted advertising delivery density, thereby improving the retention rate of the new user account on the platform and reducing the loss rate of the new user account on the platform.

[0194] Other embodiments of the present disclosure will readily occur to those skilled in the art after considering the specification and practicing the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of the present disclosure that follow the general principles of the present disclosure and include common knowledge or customary techniques in the art not disclosed herein. The description and examples are to be considered as exemplary only, with the true scope and spirit of the present disclosure being indicated by the following claims.

[0195] It should be understood that the present disclosure is not limited to the exact structures that have been described above and shown in the drawings, and that various modifications and changes can be made without departing from the scope thereof. The scope of the present disclosure is limited only by the appended claims.

Claims

1. An advertisement delivery method, characterized in that: The method comprises: When it is detected that the target user account is a first type of user account, a target user feature of the target user account is obtained, wherein the first type of user account is a user account whose registration time is less than a preset time; Determining a target retention rate of the target user account according to the target user characteristics; When the target retention rate is less than the preset retention rate, adjusting the advertisement delivery density of the target user account, and delivering advertisements to the target user account through the adjusted advertisement delivery density, wherein the preset retention rate is determined according to the retention rate of the second type of user accounts, and the second type of user accounts are user accounts with a registration duration greater than the preset duration; Wherein, adjusting the advertisement delivery density of the target user account includes: Obtain the initial advertising density of the target user account; Determining an advertisement delivery density adjustment coefficient of the target user account according to the target retention rate, the preset retention rate, and the registration duration of the target user account; The initial advertisement delivery density is adjusted by using the advertisement delivery density adjustment coefficient to obtain an adjusted advertisement delivery density; When the difference between the target retention rate and the preset retention rate is the same, the longer the registration time of the target user account is, the greater the advertisement delivery density adjustment coefficient of the target user account is; When the registration duration of the target user accounts is the same, the greater the difference between the target retention rate and the preset retention rate, the smaller the advertising delivery density adjustment coefficient of the target user accounts.

2. The method according to claim 1, characterized in that The target user feature is a target user behavior feature corresponding to the target user account, and determining the target retention rate of the target user account by using the target user feature includes: A behavior feature vector of the target user behavior feature is extracted, and a target retention rate of the target user account is determined based on the similarity between the extracted behavior feature vector and the historical behavior feature vector of the historical user behavior feature corresponding to the historical first category user account.

3. The method according to claim 1, characterized in that The target user feature is a target user portrait feature corresponding to the target user account, and determining the target retention rate of the target user account by using the target user feature includes: A portrait feature vector of the target user portrait feature is extracted, and a target retention rate of the target user account is determined based on the similarity between the extracted portrait feature vector and the historical portrait feature vector of the historical user portrait feature corresponding to the historical first category user account.

4. The method according to claim 1, characterized in that: The target user characteristics are target user behavior characteristics and target user portrait characteristics corresponding to the target user account, and determining the target retention rate of the target user account by using the target user characteristics includes: A target feature vector of the target user behavior feature and the target user portrait feature is extracted, and a target retention rate of the target user account is determined based on the similarity between the extracted feature vector and the historical feature vectors of the historical user behavior feature and the historical user portrait feature corresponding to the historical first category user account.

5. The method according to claim 1, characterized in that: The determining, according to the target retention rate, the preset retention rate and the registration duration of the target user account, the advertisement delivery density adjustment coefficient of the target user account includes: The advertisement delivery density adjustment coefficient of the target user account is calculated according to the following formula: in, is the advertising adjustment factor, is the registration duration of the target user account, is the target retention rate, is the preset retention rate.

6. The method according to claim 1, characterized in that The initial advertisement delivery density is obtained in the following manner: Initial advertisement delivery density = the number of initially delivered advertisements / (the number of initially delivered advertisements + the number of initially delivered multimedia information).

7. An advertisement delivery device, characterized in that: The device comprises: The user feature acquisition module is configured to acquire the target user features of the target user account when it is detected that the target user account is a first type of user account, wherein the first type of user account is a user account whose registration time is less than a preset time; the retention rate determination module is configured to determine the target retention rate of the target user account through the target user features; an advertisement delivery density adjustment module, configured to adjust the advertisement delivery density of the target user account when the target retention rate is less than a preset retention rate, and deliver advertisements to the target user account through the adjusted advertisement delivery density, wherein the preset retention rate is determined according to the retention rate of the second type of user account, and the second type of user account is a user account whose registration duration is greater than the preset duration; Wherein, the advertisement delivery density adjustment module includes: An initial advertisement delivery density acquisition unit is configured to acquire an initial advertisement delivery density of a target user account; an advertisement delivery density adjustment coefficient determination unit, configured to determine the advertisement delivery density adjustment coefficient of the target user account according to the target retention rate, the preset retention rate and the registration duration of the target user account; An adjusted advertisement delivery density obtaining unit is configured to adjust the initial advertisement delivery density by using the advertisement delivery density adjustment coefficient to obtain an adjusted advertisement delivery density; When the difference between the target retention rate and the preset retention rate is the same, the longer the registration time of the target user account is, the greater the advertisement delivery density adjustment coefficient of the target user account is; When the registration duration of the target user accounts is the same, the greater the difference between the target retention rate and the preset retention rate, the smaller the advertising delivery density adjustment coefficient of the target user accounts.

8. The device according to claim 7, characterized in that The target user feature is a target user behavior feature corresponding to the target user account. The retention rate determination module is specifically configured to execute: A behavior feature vector of the target user behavior feature is extracted, and a target retention rate of the target user account is determined based on the similarity between the extracted behavior feature vector and the historical behavior feature vector of the historical user behavior feature corresponding to the historical first category user account.

9. The device according to claim 7, characterized in that The target user feature is a target user portrait feature corresponding to the target user account. The retention rate determination module is specifically configured to execute: A portrait feature vector of the target user portrait feature is extracted, and a target retention rate of the target user account is determined based on the similarity between the extracted portrait feature vector and the historical portrait feature vector of the historical user portrait feature corresponding to the historical first category user account.

10. The device according to claim 7, characterized in that The target user characteristics are the target user behavior characteristics and the target user portrait characteristics corresponding to the target user account. The retention rate determination module is specifically configured to execute: A target feature vector of the target user behavior feature and the target user portrait feature is extracted, and a target retention rate of the target user account is determined based on the similarity between the extracted feature vector and the historical feature vectors of the historical user behavior feature and the historical user portrait feature corresponding to the historical first category user account.

11. The device according to claim 7, characterized in that The advertisement delivery density adjustment coefficient determination unit is specifically configured to execute: The advertisement delivery density adjustment coefficient of the target user account is calculated according to the following formula: in, is the advertising adjustment factor, is the registration duration of the target user account, is the target retention rate, is the preset retention rate.

12. The device according to claim 7, characterized in that The initial advertisement delivery density acquisition unit is configured to obtain the initial advertisement delivery density by: Initial advertisement delivery density = the number of initially delivered advertisements / (the number of initially delivered advertisements + the number of initially delivered multimedia information).

13. An electronic device, characterized in that: include: processor; a memory for storing instructions executable by the processor; The processor is configured to execute the instructions to implement the advertisement delivery method as described in any one of claims 1 to 6.

14. A storage medium, characterized in that: When the instructions in the storage medium are executed by a processor of an electronic device, the electronic device is enabled to execute the advertisement delivery method as described in any one of claims 1 to 6.

15. A computer program product comprising instructions, characterized in that When it is run on a computer, the computer is enabled to implement the advertisement delivery method as described in any one of claims 1 to 6.

Images