Advertisement delivery method and device, computer device and storage medium

Abstract

The present disclosure provides an advertisement delivery method and device, computer equipment and a storage medium. The method comprises: for each to-be-delivered user, predicting according to user characteristics, advertisement characteristics and a preset delivery number to obtain a predicted conversion rate of each to-be-delivered user for each delivery under the preset delivery number; for a current to-be-delivered round, determining a current conversion rate gain of a user in the current to-be-delivered round based on a delivered number of each to-be-delivered user and the predicted conversion rate of each to-be-delivered user for each delivery under the preset delivery number; determining a target user from a plurality of to-be-delivered users according to the current conversion rate gain of the user in the current to-be-delivered round, and delivering an advertisement in the current round to the target user. Through the above method, adaptive matching of delivery frequency and target users can be achieved under budget constraints, thereby achieving the effect of globally optimizing advertisement conversion.

Description

Technical Field

[0001] This disclosure relates to the field of computer technology, and in particular to advertising delivery methods, apparatus, computer equipment, and storage media. Background Technology

[0002] In recent years, with the rapid development of internet technology and the popularization of digital marketing, internet advertising has become an important means for businesses to acquire users and improve conversion rates. Advertisers use precise user profiles and behavioral data analysis to push ads to potential consumers, thereby improving advertising conversion efficiency. Currently, advertising technology has gradually shifted from traditional broad coverage to intelligent targeting based on big data and machine learning, relying on technologies such as real-time bidding and programmatic buying to achieve more precise targeted advertising. Summary of the Invention

[0003] To overcome the problems existing in the related technologies, this disclosure provides an advertising delivery method, apparatus, computer equipment, and storage medium.

[0004] According to a first aspect of the present disclosure, this application provides an advertising delivery method, the method comprising:

[0005] For each user to be targeted, predictions are made based on user characteristics, ad characteristics, and a preset number of targeting attempts to obtain the predicted conversion rate for each user per targeting attempt under the preset number of targeting attempts.

[0006] For the current round of advertising, based on the number of times each user has been advertised and the predicted conversion rate of each user for each advertised campaign under the preset number of campaigns, the current conversion rate gain of the current round of advertising is determined.

[0007] Based on the current conversion rate gain of the users in the current round of advertising, target users are identified from multiple users to be advertised, and advertising in the current round is delivered to the target users.

[0008] In any embodiment of this disclosure, determining the current conversion rate gain for the current round of advertising, based on the number of times each user has been advertised and the predicted conversion rate for each user under the preset number of advertised times, includes:

[0009] For each user to be targeted, the difference between the predicted conversion rate of the user corresponding to the current number of times the campaign has been conducted and the predicted conversion rate corresponding to the next campaign after the current number of times the campaign has been conducted is used as the current conversion rate gain of the user in the current campaign round.

[0010] In any embodiment of this disclosure, determining the target user from a plurality of users to be targeted based on the current conversion rate gain of each user to be targeted includes:

[0011] Obtain a preset number of the largest current conversion rate gains among the current conversion rate gains of the users in the current round of advertising, and determine the users in the current round of advertising corresponding to the preset number of largest current conversion rate gains as the target users of the current round of advertising.

[0012] In conjunction with any embodiment of this disclosure, the conversion rate prediction model is trained through the following process:

[0013] The sample training data in the sample training dataset is input into the original model to obtain the predicted conversion rate of each sample user for each delivery under the preset number of deliveries of the sample model. The sample training data includes sample user features, sample ad features and the preset number of deliveries of the sample. Each sample training data is labeled with the actual conversion tag of each sample user for each delivery under the preset number of deliveries of the sample.

[0014] Based on the predicted conversion rate of each sample user for each delivery under the preset number of deliveries, and the actual conversion tag of each sample user for each delivery under the preset number of deliveries, the parameters of the original model are tuned to obtain a trained conversion rate prediction model.

[0015] In any embodiment of this disclosure, the actual conversion tag for each campaign targeting the sample user before a conversion occurs is "not converted"; and the actual conversion tag for each campaign targeting the user after a conversion occurs is "converted".

[0016] In conjunction with any embodiment of this disclosure, the method further includes:

[0017] Obtain the actual conversion tag for each user to be targeted after each delivery under the preset number of deliveries;

[0018] Based on the predicted conversion rate of each target user for each delivery under the preset number of deliveries, and the actual conversion tag of each target user after each delivery under the preset number of deliveries, the parameters of the conversion rate prediction model are updated to obtain the updated conversion rate prediction model.

[0019] In any embodiment of this disclosure, the user characteristics of each user to be targeted include the user's attribute characteristics and historical behavior characteristics.

[0020] Secondly, this application also provides an advertising delivery device, comprising:

[0021] The conversion rate prediction module is used to predict the conversion rate of each user to be targeted based on user characteristics, ad characteristics and preset number of targeting times, and to obtain the predicted conversion rate of each user to be targeted for each targeting time under the preset number of targeting times.

[0022] The gain determination module is used to determine the current conversion rate gain of the user in the current round of advertising based on the number of times each user has been advertised and the predicted conversion rate of each user under the preset number of advertising times.

[0023] The advertising delivery module is used to determine the target user from multiple users to be advertised based on the current conversion rate gain of the users in the current round of advertising, and to deliver the current round of advertising to the target user.

[0024] In one embodiment, the gain determination module is specifically used for:

[0025] For each user to be targeted, the difference between the predicted conversion rate of the user corresponding to the current number of times the campaign has been conducted and the predicted conversion rate corresponding to the next campaign after the current number of times the campaign has been conducted is used as the current conversion rate gain of the user in the current campaign round.

[0026] In one embodiment, the advertising delivery module is specifically used for:

[0027] Obtain a preset number of the largest current conversion rate gains among the current conversion rate gains of the users in the current round of advertising, and determine the users in the current round of advertising corresponding to the preset number of largest current conversion rate gains as the target users of the current round of advertising.

[0028] In one embodiment, the advertising delivery device further includes a model training module, which is used for:

[0029] The sample training data in the sample training dataset is input into the original model to obtain the predicted conversion rate of each sample user for each delivery under the preset number of deliveries of the sample model. The sample training data includes sample user features, sample ad features and the preset number of deliveries of the sample. Each sample training data is labeled with the actual conversion tag of each sample user for each delivery under the preset number of deliveries of the sample.

[0030] Based on the predicted conversion rate of each sample user for each delivery under the preset number of deliveries, and the actual conversion tag of each sample user for each delivery under the preset number of deliveries, the parameters of the original model are tuned to obtain a trained conversion rate prediction model.

[0031] In one embodiment, the actual conversion tag for each campaign to the sample user before a conversion is generated is "not converted"; and the actual conversion tag for each campaign to the time and thereafter after a conversion is "converted".

[0032] In one embodiment, the advertising delivery device further includes a model update module, which is used to:

[0033] Obtain the actual conversion tag for each user to be targeted after each delivery under the preset number of deliveries;

[0034] Based on the predicted conversion rate of each target user for each delivery under the preset number of deliveries, and the actual conversion tag of each target user after each delivery under the preset number of deliveries, the parameters of the conversion rate prediction model are updated to obtain the updated conversion rate prediction model.

[0035] In one embodiment, the user characteristics of each user to be targeted include the user's attribute characteristics and historical behavior characteristics.

[0036] Thirdly, this application provides a computer program product, including a computer program / instructions, which, when executed by a processor, implement the steps of the method described in any embodiment.

[0037] Fourthly, this application provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the steps of the method described in any of the above embodiments.

[0038] Fifthly, this application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method described in any of the above embodiments.

[0039] The technical solutions provided by the embodiments of this disclosure may include the following beneficial effects:

[0040] Compared to current advertising campaigns that only consider the effect of a single campaign, this embodiment predicts the conversion rate of each user in a preset number of campaigns. Then, for the current campaign round, based on the number of campaigns already run for each user and the predicted conversion rate of each campaign within the preset number of campaigns, it can more accurately and reasonably calculate the conversion rate gain for each user in the current campaign round. Finally, based on the conversion rate gain of each user in the current campaign round, target users are identified from multiple users, and ads are delivered to these target users. This allows for adaptive matching of campaign frequency and target users within budget constraints (i.e., a limited total number of campaigns), avoiding over-targeting of inefficient users while ensuring sufficient reach to high-potential users, thereby achieving a globally optimized advertising conversion effect.

[0041] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit this disclosure. Attached Figure Description

[0042] The accompanying drawings, which are incorporated in and form part of this disclosure, illustrate embodiments consistent with this disclosure and, together with the description, serve to explain the principles of this disclosure.

[0043] Figure 1 These are application scenario diagrams illustrating an advertising delivery method through some exemplary embodiments.

[0044] Figure 2 These are schematic diagrams illustrating one method of determining conversion gain, as shown in some exemplary embodiments.

[0045] Figure 3 These are schematic diagrams illustrating one method of identifying a target user, as shown in some exemplary embodiments.

[0046] Figure 4 This is a flowchart illustrating an advertising delivery method using some exemplary embodiments.

[0047] Figure 5 These are structural diagrams of an original model illustrating some exemplary embodiments.

[0048] Figure 6 This is a flowchart illustrating another advertising delivery method using some exemplary embodiments.

[0049] Figure 7 This is a block diagram illustrating another advertising delivery device using some exemplary embodiments.

[0050] Figure 8 These are hardware structure diagrams of a computer device illustrating some exemplary embodiments. Detailed Implementation

[0051] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numerals in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this disclosure. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this disclosure as detailed in the appended claims.

[0052] The terminology used in this disclosure is for the purpose of describing particular embodiments only and is not intended to be limiting of the disclosure. The singular forms “a,” “the,” and “the” as used in this disclosure and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used herein refers to and includes any and all possible combinations of one or more of the associated listed items.

[0053] It should be understood that although the terms first, second, third, etc., may be used in this disclosure to describe various information, such information should not be limited to these terms. These terms are used only to distinguish information of the same type from one another. For example, without departing from the scope of this disclosure, first information may also be referred to as second information, and similarly, second information may also be referred to as first information. Depending on the context, the word "if" as used herein may be interpreted as "when," "when," or "in response to determination."

[0054] In real-world scenarios, different users respond significantly differently to ads: some users may make a purchase after the first exposure, while others require multiple exposures to convert, or even fail to convert at all. Relying solely on the results of a single ad campaign can lead to uneven budget allocation, such as over-targeting low-value users or under-targeting high-potential users, thus reducing overall advertising efficiency. Related technologies primarily rely on predicting click-through rates or conversion rates for single exposures (e.g., using logistic regression models) to determine ad placement. However, in practice, this approach only considers the conversion probability of each user in a single ad exposure, neglecting the need for multiple exposures for some users to convert. This single-shot decision-making mechanism, lacking a holistic perspective, not only causes budget imbalances but also misses conversion opportunities that could be achieved through continuous nurturing, resulting in low conversion prediction accuracy and poor overall optimization capabilities.

[0055] In view of this, the present disclosure provides an advertising delivery method, apparatus, computer equipment, and storage medium. The method can be used in business scenarios that require multiple advertising deliveries to users (such as e-commerce promotions, application installations, etc.).

[0056] Optionally, the method can be executed by an advertising platform, such as an e-commerce platform, or by a server, electronic device, etc.

[0057] The embodiments of this disclosure will now be described in detail.

[0058] This disclosure provides, in its first aspect, an advertising delivery method. Please refer to [link / reference needed]. Figure 1 It includes the following steps:

[0059] S101, for each user to be targeted, make a prediction based on user characteristics, advertising characteristics and preset number of targeting times, and obtain the predicted conversion rate of each user for each targeting time under the preset number of targeting times.

[0060] Here, "users to be targeted" refers to users who require current advertising. User characteristics include attribute features and historical behavioral features. Attribute features include basic characteristics such as name and gender, while historical behavioral features include search and purchase behaviors over historical periods. Advertising features refer to the relevant characteristics of the advertisements to be targeted, such as ad format, content, and placement. The preset number of deliveries refers to a pre-set fixed number of ad deliveries to the target users, such as 10 deliveries. The predicted conversion rate refers to the predicted conversion rate for each target user across the preset 10 deliveries. For example, the predicted conversion rate for the 5th ad delivery within the preset 10 deliveries is essentially the predicted conversion rate obtained by delivering 5 additional ads to the target user within the fixed 10 deliveries.

[0061] S102, for the current round of delivery, based on the number of deliveries for each user to be delivered and the predicted conversion rate for each user to be delivered under the preset number of deliveries, determine the current conversion rate gain for the current round of delivery.

[0062] The number of times an ad has been placed for a user to be targeted refers to the number of times an ad has been placed for that user before the current round of targeting. For example, if the current round of targeting is the 5th round, and the user has been targeted 2 times in the previous 4 rounds, then the current number of times an ad has been placed for that user is 2.

[0063] Optionally, the number of times each user has been targeted before the current targeting round can be determined. Then, for the current targeting round, the difference between the predicted conversion rate of each user corresponding to the number of times they have been targeted and the predicted conversion rate corresponding to the next targeting round after the number of times they have been targeted can be used as the current conversion rate gain of the user in the current targeting round. This can more accurately and reasonably determine the current conversion rate gain of each user in the current targeting round.

[0064] For example, if the current campaign round is the 5th campaign, the number of times each user has been campaigned before the current campaign round can be determined. If any user has been campaigned twice before the current campaign round, the difference between the predicted conversion rate of the user in the 2nd campaign and the predicted conversion rate in the 3rd campaign can be used as the current conversion rate gain for that user in the current campaign round (i.e., the 5th campaign). If any user has been campaigned zero times before the current campaign round, the predicted conversion rate of the user in the 1st campaign can be directly used as the current conversion rate gain for that user in the current campaign round (i.e., the 5th campaign).

[0065] S103, based on the current conversion rate gain of the users in the current round of advertising, determine the target user from multiple users to be advertised, and deliver the current round of advertising to the target user.

[0066] Optionally, after determining the current conversion rate gain of each user to be targeted, a preset number of the largest current conversion rate gains among the current conversion rate gains of the users in the current round of targeting can be obtained. The users in the current round of targeting corresponding to the preset number of largest current conversion rate gains are identified as the target users of the current round of targeting. This allows for more accurate identification of the target users for each targeting session and more precise targeting of ads to the target users, thereby maximizing the effectiveness of the targeting in real time.

[0067] For example, the current conversion rate gains of the users in the current round of advertising can be sorted in descending order, and the users in the current round of advertising corresponding to the largest current conversion rate gains of a preset number can be identified as target users. For example, if the preset number is 3 users, the users corresponding to the 3 largest current conversion rate gains of the users in the current round of advertising can be identified as target users from the descending sequence of the current conversion rate gains of the multiple users, and the advertising can be carried out on these 3 target users.

[0068] For example, after determining the predicted conversion rate for each user to be targeted to in each campaign within a preset number of campaigns, the conversion rate gain for each user to be targeted to in each campaign within the preset number of campaigns can be determined. For example... Figure 2 As shown, the users to be targeted include users A, B, and C. User A's predicted conversion rate for the first targeting is 0.1, for the second targeting is 0.3, and for the third targeting is 0.35. User B's predicted conversion rate for the first targeting is 0.05, for the second targeting is 0.2, and for the third targeting is 0.25. User C's predicted conversion rate for the first targeting is 0.3, for the second targeting is 0.31, and for the third targeting is... The conversion rate is 0.32; therefore, it can be determined that User A's conversion rate gain in the first campaign is 0.1, User A's conversion rate gain in the second campaign is 0.2, and User A's conversion rate gain in the third campaign is 0.05; User B's conversion rate gain in the first campaign is 0.05, User B's conversion rate gain in the second campaign is 0.15, and User B's conversion rate gain in the third campaign is 0.05; User C's conversion rate gain in the first campaign is 0.3, User C's conversion rate gain in the second campaign is 0.01, and User C's conversion rate gain in the third campaign is 0.01.

[0069] Furthermore, the target users for each campaign can be determined based on the conversion rate gains of users A, B, and C in each campaign. For example... Figure 3 As shown, in the first campaign, 0.1, 0.05, and 0.3 can be compared to determine that user C, corresponding to 0.3, is the target user. In the second campaign, since user C has already been campaigned once, while users A and B have not been campaigned, 0.1, 0.05, and 0.01 can be compared to determine that user A, corresponding to 0.1, is the target user. In the third campaign, since users A and C have both been campaigned once, while user B has not been campaigned, 0.2, 0.05, and 0.01 can be compared to determine that user A, corresponding to 0.2, is the target customer.

[0070] It should be noted that in actual use, S101 only needs to be executed once to obtain the predicted conversion rate of each user to be targeted for each campaign under the preset number of campaigns; while S102 and S103 need to be executed once during each campaign under the preset number of campaigns in order to determine the target users for each campaign.

[0071] Compared to current advertising campaigns that only consider the effect of a single campaign, this embodiment predicts the conversion rate of each user in a preset number of campaigns. Then, for the current campaign round, based on the number of campaigns already run for each user and the predicted conversion rate of each campaign within the preset number of campaigns, it can more accurately and reasonably calculate the conversion rate gain for each user in the current campaign round. Finally, based on the conversion rate gain of each user in the current campaign round, target users are identified from multiple users, and ads are delivered to these target users. This allows for adaptive matching of campaign frequency and target users within budget constraints (i.e., a limited total number of campaigns), avoiding over-targeting of inefficient users while ensuring sufficient reach to high-potential users, thereby achieving a globally optimized advertising conversion effect.

[0072] In one exemplary embodiment, such as Figure 4 As shown, the above conversion rate prediction model provides a model training method, which specifically includes the following steps:

[0073] S401, input the sample training data in the sample training dataset into the original model to obtain the predicted conversion rate of each sample user for each delivery under the preset number of deliveries output by the original model.

[0074] The sample training data includes sample user features, sample ad features, and the preset number of ad placements. Sample users can be users who have received ad placements within a historical time period. Sample user features include attribute features and historical behavioral features. Each sample training data point is labeled with the actual conversion tag for each sample user for each placement within the preset number of placements. The original model architecture can be as follows: Figure 5 As shown, the underlying architecture uses a traditional deep learning model, and the intermediate layer uses an MLP (Multilayer Perceptron) for feature processing.

[0075] Optionally, the sample training data from the pre-collected sample training dataset can be input into the original model, and the predicted conversion rate for each sample user in each delivery under the preset number of deliveries can be determined through the original model.

[0076] S402, based on the predicted conversion rate of each sample user for each delivery under the preset number of deliveries, and the actual conversion tag of each sample user for each delivery under the preset number of deliveries, the parameters of the original model are adjusted to obtain a trained conversion rate prediction model.

[0077] The actual conversion tag of the sample user after each campaign is used to characterize the actual conversion behavior of the sample user after each campaign. It can include no conversion and converted (or 0 and 1, etc.). The actual conversion tag of the sample user before a conversion is generated is no conversion; the actual conversion tag of the campaign after a conversion is generated is converted.

[0078] Optionally, after obtaining the predicted conversion rate for each sample user under the preset number of deliveries, the difference between the predicted conversion rate for each sample user under the preset number of deliveries and the actual conversion tag for each user after each delivery can be compared. By calculating the loss function (such as cross-entropy loss value) between the predicted conversion rate and the actual tag, the model parameters of the original model can be adjusted according to the loss function. After multiple iterations of training, the finally trained conversion rate prediction model is obtained.

[0079] In this embodiment, by inputting sample data containing sample user characteristics, sample ad characteristics, and a preset number of ad placements into the model for training, the model can accurately learn the changing patterns of user conversion behavior under different ad placement frequencies. Based on parameter tuning by comparing the predicted conversion rate with the actual conversion tags, the model's accuracy in depicting user response patterns can be optimized, thereby constructing an intelligent model that can dynamically predict conversion effects at each stage of ad placement. This training process effectively captures the conversion probability curve of users from initial contact to multiple repeated exposures, providing a scientific basis for subsequent ad placement decisions.

[0080] Based on the above embodiments, in an exemplary embodiment, such as Figure 6 As shown, the method also provides a model update approach, which specifically includes the following steps:

[0081] S601, Obtain the actual conversion tag of each user to be targeted after each targeting session under the preset targeting number of times.

[0082] Optionally, the actual conversion results of each user to be targeted can be obtained for each ad campaign, and the actual conversion tags of each user after each campaign with a preset number of campaigns can be obtained.

[0083] S602, based on the predicted conversion rate of each target user for each delivery under the preset number of deliveries, and the actual conversion tag of each target user after each delivery under the preset number of deliveries, the parameters of the conversion rate prediction model are updated to obtain the updated conversion rate prediction model.

[0084] Optionally, the difference between the predicted conversion rate of each user to be targeted and the actual conversion tag of each user after each targeting under the preset number of targeting can be compared. By calculating the loss function between the predicted conversion rate and the actual tag, the model parameters of the conversion rate prediction model can be updated according to the loss function. After multiple iterations of training, the updated conversion rate prediction model is finally obtained.

[0085] In this embodiment, by continuously collecting the latest user campaign data and actual conversion results, and comparing the actual conversion results with the predicted conversion rate, the conversion rate prediction model can be updated and iterated. This enables the conversion rate prediction model to always remain sensitive to market changes and user behavior trends, continuously improve the model's recognition accuracy of various user response patterns, and achieve a virtuous cycle in which the campaign effect continues to improve over time.

[0086] Corresponding to the embodiments of the foregoing methods, this disclosure also provides embodiments of the apparatus and the terminal to which it is applied.

[0087] Secondly, this application also provides an advertising delivery device, which can be used in business scenarios that require multiple advertising deliveries to users (such as e-commerce promotions, application installations, etc.). The device can be applied to advertising delivery platforms, such as e-commerce platforms, or servers, electronic devices, etc.

[0088] like Figure 7 As shown, the advertising delivery device includes:

[0089] The conversion rate prediction module 701 is used to predict the conversion rate of each user to be targeted based on user characteristics, advertising characteristics and preset number of targeting times, so as to obtain the predicted conversion rate of each user to be targeted for each targeting time under the preset number of targeting times.

[0090] The gain determination module 702 is used to determine the current conversion rate gain of the user in the current round of delivery based on the number of deliveries for each user and the predicted conversion rate of each user for each delivery under the preset number of deliveries for the current round of delivery.

[0091] The advertising delivery module 703 is used to determine the target user from multiple users to be delivered to based on the current conversion rate gain of the users in the current round of advertising, and to deliver the current round of advertising to the target user.

[0092] In one embodiment, the gain determination module 702 is specifically used for:

[0093] For each user to be targeted, the difference between the predicted conversion rate of the user corresponding to the current number of times the campaign has been conducted and the predicted conversion rate corresponding to the next campaign after the current number of times the campaign has been conducted is used as the current conversion rate gain of the user in the current campaign round.

[0094] In one embodiment, the advertising delivery module 703 is specifically used for:

[0095] Obtain a preset number of the largest current conversion rate gains among the current conversion rate gains of the users in the current round of advertising, and determine the users in the current round of advertising corresponding to the preset number of largest current conversion rate gains as the target users of the current round of advertising.

[0096] In one embodiment, the advertising delivery device further includes a model training module, which is used for:

[0097] The sample training data in the sample training dataset is input into the original model to obtain the predicted conversion rate of each sample user for each delivery under the preset number of deliveries of the sample model. The sample training data includes sample user features, sample ad features and the preset number of deliveries of the sample. Each sample training data is labeled with the actual conversion tag of each sample user for each delivery under the preset number of deliveries of the sample.

[0098] Based on the predicted conversion rate of each sample user for each delivery under the preset number of deliveries, and the actual conversion tag of each sample user for each delivery under the preset number of deliveries, the parameters of the original model are tuned to obtain a trained conversion rate prediction model.

[0099] In one embodiment, the actual conversion tag for each campaign to the sample user before a conversion is generated is "not converted"; and the actual conversion tag for each campaign to the time and thereafter after a conversion is "converted".

[0100] In one embodiment, the advertising delivery device further includes a model update module, which is used to:

[0101] Obtain the actual conversion tag for each user to be targeted after each delivery under the preset number of deliveries;

[0102] Based on the predicted conversion rate of each target user for each delivery under the preset number of deliveries, and the actual conversion tag of each target user after each delivery under the preset number of deliveries, the parameters of the conversion rate prediction model are updated to obtain the updated conversion rate prediction model.

[0103] In one embodiment, the user characteristics of each user to be targeted include the user's attribute characteristics and historical behavior characteristics.

[0104] The specific implementation process of the functions and roles of each module in the above device can be found in the implementation process of the corresponding steps in the above method, and will not be repeated here.

[0105] A third aspect of this disclosure provides a computer program product including a computer program / instructions that, when executed by a processor, implement the method described in the first aspect.

[0106] For the device embodiments and computer program product embodiments, since they basically correspond to the method embodiments, the relevant parts can be referred to in the description of the method embodiments. Furthermore, the device embodiments described above are merely illustrative; the modules described as separate components may or may not be physically separate, and the components shown as modules may or may not be physical modules, i.e., they may be located in one place or distributed across multiple network modules. Some or all of the modules can be selected to achieve the purpose of this disclosure according to actual needs. Those skilled in the art can understand and implement this without any inventive effort.

[0107] Fourthly, the embodiments of the advertising delivery device provided in this disclosure can be applied to computer equipment. The computer equipment can be used in business scenarios requiring multiple advertising deliveries to users (such as e-commerce promotions, application installations, etc.). The computer equipment may include an advertising delivery platform, such as an e-commerce platform, or a server, electronic equipment, etc.

[0108] Please refer to the appendix. Figure 8 The illustration exemplifies a hardware schematic of a computer device. For example, device 800 could be a mobile phone, computer, digital broadcasting terminal, messaging device, game console, tablet device, medical device, fitness equipment, personal digital assistant, etc.

[0109] Device 800 may include one or more of the following components: processing component 801, memory 802, power supply component 803, multimedia component 804, audio component 805, input / output (I / O) interface 806, sensor component 807, and communication component 808.

[0110] Processing component 801 typically controls the overall operation of device 800, such as actions associated with display, telephone calls, data communication, camera actions, and recording actions. Processing component 801 may include one or more processors 809 to execute instructions to complete all or part of the steps of the methods described above. Furthermore, processing component 801 may include one or more modules to facilitate interaction between processing component 801 and other components. For example, processing component 801 may include a multimedia module to facilitate interaction between multimedia component 804 and processing component 801.

[0111] Memory 802 is configured to store various types of data to support the operation of device 800. Examples of this data include instructions for any application or method operating on device 800, contact data, phonebook data, messages, pictures, videos, etc. Memory 802 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 storage, flash memory, magnetic disk, or optical disk.

[0112] The power supply component 803 provides power to the various components of the device 800. The power supply component 803 may include a power management system, one or more power supplies, and other components associated with generating, managing, and distributing power to the device 800.

[0113] Multimedia component 804 includes a screen that provides an output interface between the device 800 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 touchscreen to receive input signals from the user. The touch panel includes one or more touch sensors to sense touch, swipe, and gestures on the touch panel. The touch sensors may sense not only the boundaries of the touch or swipe action but also the duration and pressure associated with the touch or swipe action. In some embodiments, multimedia component 804 includes a front-facing camera and / or a rear-facing camera. When the device 800 is in an active mode, such as a shooting mode or a video mode, the front-facing camera and / or the rear-facing camera may receive external multimedia data. Each front-facing camera and rear-facing camera may be a fixed optical lens system or have focal length and optical zoom capabilities.

[0114] Audio component 805 is configured to output and / or input audio signals. For example, audio component 805 includes a microphone (MIC) configured to receive external audio signals when device 800 is in an operational mode, such as call mode, recording mode, and voice recognition mode. The received audio signals may be further stored in memory 802 or transmitted via communication component 808. In some embodiments, audio component 805 also includes a speaker for outputting audio signals.

[0115] I / O interface 806 provides an interface between processing component 801 and peripheral interface modules, such as keyboards, click wheels, buttons, etc. These buttons may include, but are not limited to, home buttons, volume buttons, power buttons, and lock buttons.

[0116] Sensor assembly 807 includes one or more sensors for providing status assessments of various aspects of device 800. For example, sensor assembly 807 can detect the on / off state of device 800, the relative positioning of components such as the display and keypad of device 800, changes in the position of device 800 or a component of device 800, the presence or absence of user contact with device 800, the orientation or acceleration / deceleration of device 800, and temperature changes of device 800. Sensor assembly 807 may also include a proximity sensor configured to detect the presence of nearby objects without any physical contact. Sensor assembly 807 may also include a light sensor, such as a CMOS or CCD image sensor, for use in imaging applications. In some embodiments, sensor assembly 807 may also include an accelerometer, a gyroscope, a magnetometer, a pressure sensor, or a temperature sensor.

[0117] Communication component 808 is configured to facilitate wired or wireless communication between device 800 and other devices. Device 800 can access wireless networks based on communication standards, such as WiFi, 2G or 3G, 4G or 5G, or combinations thereof. In one exemplary embodiment, communication component 808 receives broadcast signals or broadcast-related information from an external broadcast management system via a broadcast channel. In one exemplary embodiment, communication component 808 also includes a near-field communication (NFC) module to facilitate short-range communication. For example, the NFC module may be implemented based on radio frequency identification (RFID) technology, Infrared Data Association (IrDA) technology, ultra-wideband (UWB) technology, Bluetooth (BT) technology, and other technologies.

[0118] In an exemplary embodiment, device 800 may 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 perform the advertising delivery method of the computer device described above.

[0119] Fifthly, in exemplary embodiments, this disclosure also provides a non-transitory computer-readable storage medium including instructions, such as a memory 802 including instructions, which can be executed by a processor 809 of device 800 to complete the advertising delivery method of the computer device. For example, the non-transitory computer-readable storage medium may be a ROM, random access memory (RAM), CD-ROM, magnetic tape, floppy disk, and optical data storage device, etc.

[0120] The foregoing has described specific embodiments of this disclosure. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims may be performed in a different order than that shown in the embodiments and may still achieve the desired results. Furthermore, the processes depicted in the drawings do not necessarily require the specific or sequential order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0121] Other embodiments of this disclosure will readily occur to those skilled in the art upon consideration of the specification and practice of the invention applied herein. This disclosure is intended to cover any variations, uses, or adaptations of this disclosure that follow the general principles of this disclosure and include common knowledge or customary techniques in the art not claimed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of this disclosure are indicated by the following claims.

[0122] It should be understood that this disclosure is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this disclosure is limited only by the appended claims.

[0123] The above description is merely a preferred embodiment of this disclosure and is not intended to limit this disclosure. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this disclosure should be included within the scope of protection of this disclosure.

Claims

1. An advertising placement method, characterized in that, The method includes: For each user to be targeted, predictions are made based on user characteristics, ad characteristics, and a preset number of targeting attempts to obtain the predicted conversion rate for each user per targeting attempt under the preset number of targeting attempts. For the current round of advertising, based on the number of times each user has been advertised and the predicted conversion rate of each user for each advertised campaign under the preset number of campaigns, the current conversion rate gain of the current round of advertising is determined. Based on the current conversion rate gain of the users in the current round of advertising, target users are identified from multiple users to be advertised, and advertising in the current round is delivered to the target users.

2. The method according to claim 1, characterized in that, The step of determining the current conversion rate gain for the current round of advertising, based on the number of times each user has been advertised and the predicted conversion rate for each user under the preset number of advertised times, includes: For the current round of advertising, the difference between the predicted conversion rate of each user in the previous number of advertising sessions and the predicted conversion rate of the next advertising session in the previous number of advertising sessions is used as the current conversion rate gain for the user in the current round of advertising.

3. The method according to claim 1, characterized in that, The step of determining the target user from multiple users to be targeted based on the current conversion rate gain of the users in the current round of targeting includes: Obtain a preset number of the largest current conversion rate gains among the current conversion rate gains of the users in the current round of advertising, and determine the users in the current round of advertising corresponding to the preset number of largest current conversion rate gains as the target users of the current round of advertising.

4. The method according to claim 1, characterized in that, The conversion rate prediction model is trained through the following process: The sample training data in the sample training dataset is input into the original model to obtain the predicted conversion rate of each sample user for each delivery under the preset number of deliveries of the sample model. The sample training data includes sample user features, sample ad features and the preset number of deliveries of the sample. Each sample training data is labeled with the actual conversion tag of each sample user for each delivery under the preset number of deliveries of the sample. Based on the predicted conversion rate of each sample user for each of the preset number of deliveries in the sample output by the original model, and the actual conversion tag of each sample user for each of the preset number of deliveries in the sample, the parameters of the original model are tuned to obtain a trained conversion rate prediction model.

5. The method according to claim 4, characterized in that, For the sample users, the actual conversion tag for each campaign before a conversion was generated was "not converted"; for the campaign that generated a conversion and every subsequent campaign, the actual conversion tag was "converted".

6. The method according to claim 1, characterized in that, The method further includes: Obtain the actual conversion tag for each user to be targeted after each delivery under the preset number of deliveries; Based on the predicted conversion rate of each target user for each delivery under the preset number of deliveries, and the actual conversion tag of each target user after each delivery under the preset number of deliveries, the parameters of the conversion rate prediction model are updated to obtain the updated conversion rate prediction model.

7. The method according to claim 1, characterized in that, The user characteristics of each user to be targeted include the user's attribute characteristics and historical behavior characteristics.

8. An advertising delivery device, characterized in that, The device includes: The conversion rate prediction module is used to predict the conversion rate of each user to be targeted based on user characteristics, ad characteristics and preset number of targeting times. The gain determination module is used to determine the current conversion rate gain of the user in the current round of advertising based on the number of times each user has been advertised and the predicted conversion rate of each user under the preset number of advertising times. The advertising delivery module is used to determine the target user from multiple users to be advertised based on the current conversion rate gain of the users in the current round of advertising, and to deliver the current round of advertising to the target user.

9. The apparatus according to claim 8, characterized in that, The gain determination module is specifically used for: For the current round of advertising, the difference between the predicted conversion rate of each user in the previous number of advertising sessions and the predicted conversion rate of the next advertising session in the previous number of advertising sessions is used as the current conversion rate gain for the user in the current round of advertising.

10. The apparatus according to claim 8, characterized in that, The advertising delivery module is specifically used for: Obtain a preset number of the largest current conversion rate gains among the current conversion rate gains of the users in the current round of advertising, and determine the users in the current round of advertising corresponding to the preset number of largest current conversion rate gains as the target users of the current round of advertising.

11. The apparatus according to claim 8, characterized in that, The device further includes: The model training module is used to input the sample training data from the sample training dataset into the original model to obtain the predicted conversion rate of each sample user for each delivery under the preset number of deliveries. The sample training data includes sample user features, sample ad features, and the preset number of deliveries. Each sample training data is labeled with the actual conversion tag of each sample user for each delivery under the preset number of deliveries. Based on the predicted conversion rate of each sample user for each delivery under the preset number of deliveries and the actual conversion tag of each sample user for each delivery under the preset number of deliveries, the original model is tuned to obtain a trained conversion rate prediction model.

12. The apparatus according to claim 8, characterized in that, The device further includes: The model update module is used to obtain the actual conversion tag of each user to be targeted after each deployment under the preset number of deployments; and to update the parameters of the conversion rate prediction model according to the predicted conversion rate of each target user under the preset number of deployments and the actual conversion tag of each target user after each deployment under the preset number of deployments, so as to obtain the updated conversion rate prediction model.

13. A computer program product comprising a computer program / instructions, characterized in that, When the computer program / instructions are executed by the processor, they implement the method described in any one of claims 1 to 7.

14. A computer device, characterized in that, It includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the program, implements the method as described in any one of claims 1 to 7.

15. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the method as described in any one of claims 1 to 7.

Images