An advertisement pushing method and device, a server, a client and a storage medium

Abstract

Embodiments of the present application provide an advertisement pushing method and device, a server, a client and a storage medium. The server can acquire multiple advertisements and send the multiple advertisements to the client when an advertisement pushing condition is met. The client can display preview information of the multiple advertisements on a playing interface, and then play an advertisement selected by a user based on the preview information. Since the preview information displayed on the playing interface includes the time length of the advertisement, the user can intuitively know the time length of each advertisement through the preview information. In order to shorten the time length of watching the advertisement, the user can actively select one of the advertisements. By giving the user the right to choose and taking the different benefits corresponding to different time lengths as guidance, the user's resistance to the advertisement can be reduced, the mandatory playing is changed to active selection, the user's concentration in watching the advertisement can be improved, and thus the conversion rate of the advertisement is improved.

Description

Technical Field

[0001] This application relates to the field of data processing technology, and in particular to an advertising push method, apparatus, server, client and storage medium. Background Technology

[0002] To increase product exposure, product manufacturers will advertise their products. Advertising scenarios are diverse; a common scenario is that product manufacturers place product ads on video streaming platforms, which then push these ads before or during video playback.

[0003] In current technologies, video playback platforms push advertisements to the client after the advertisement finishes playing, and the client only continues playing the video after the advertisement has finished. In this case, because the advertisement is forcibly presented to the user, the user's aversion is high, and they are likely not to focus on the advertisement content, resulting in invalid ad playback and a low click-through rate (CTR). Summary of the Invention

[0004] The purpose of this application is to provide an advertising push method, apparatus, server, client, and storage medium to improve the conversion rate of advertisements. The specific technical solution is as follows:

[0005] In a first aspect, embodiments of this application provide an advertising push method, applied to a server, the method comprising:

[0006] Multiple advertisements are obtained, wherein at least two of the advertisements have different durations, and the different durations represent the different selection rights corresponding to the user choosing different advertisements;

[0007] In response to the fulfillment of the advertising push conditions, the plurality of advertisements are sent to the client so that the client displays preview information of the plurality of advertisements on the playback interface and plays the advertisement selected by the user based on the preview information, wherein the preview information includes the duration of the advertisement.

[0008] Optionally, the step of obtaining multiple advertisements includes:

[0009] Obtain the first advertisement selected by the advertising recommendation system according to preset advertising recommendation rules;

[0010] Multiple second advertisements generated based on product characteristics and the video characteristics of the target video are obtained, wherein the target video is the video to be played or currently being played by the client, and the duration of the second advertisement is shorter than the duration of the first advertisement.

[0011] Optionally, the second advertisement can be generated in the following ways:

[0012] The video features of the target video are obtained, wherein the video features include character features, which are obtained by extracting character profiles of the characters in the target video;

[0013] Based on the degree of matching between product features and video features, the target product to be pushed with ads when playing the target video is determined from the products that need to be pushed with ads.

[0014] Based on the product characteristics of each target product and the video characteristics, generate advertising copy for each target product;

[0015] Based on the advertising copy for each target product, generate a second advertisement corresponding to each target product.

[0016] Optionally, the video features include character features and contextual features;

[0017] The step of determining the target product to be pushed an advertisement to when playing the target video from the products that need to be pushed an advertisement to, based on the degree of matching between the product characteristics and the video characteristics, includes:

[0018] For each product that needs to be pushed with advertisements, determine the first semantic relevance between the product features and the role features, and determine the second semantic relevance between the product features and the context features;

[0019] Products whose first semantic relevance is greater than a first preset threshold and whose second semantic relevance is greater than a second preset threshold are identified as target products for advertising push when playing the target video.

[0020] Optionally, the step of obtaining multiple advertisements includes:

[0021] Obtain user profiles of users using the client, wherein the user profiles reflect users' preferences for advertisements;

[0022] Based on the user profile, multiple advertisements are obtained;

[0023] After the step of sending the plurality of advertisements to the client, the method further includes:

[0024] Determine the advertisement selected by the user based on the preview information;

[0025] The user profile is updated based on the information of the advertisement selected by the user, the information of the advertisements not selected by the user among the multiple advertisements, and the video features of the target video, wherein the target video is: the video to be played or currently being played by the client.

[0026] Secondly, embodiments of this application provide an advertising push method applied to a client, the method comprising:

[0027] Receive multiple advertisements sent by the server, wherein at least two of the multiple advertisements have different durations;

[0028] The playback interface displays preview information of the multiple advertisements, wherein the preview information includes at least the duration of the advertisement, and the duration included in the preview information is used to indicate to the user the different selection benefits corresponding to different advertisements;

[0029] In response to an ad selection instruction issued by the user based on the preview information, the ad indicated by the ad selection instruction is played on the playback interface.

[0030] Optionally, the plurality of advertisements includes a first advertisement and a plurality of second advertisements, wherein the duration of the second advertisements is shorter than the duration of the first advertisements;

[0031] After the step of displaying preview information of the multiple advertisements on the playback interface, the method further includes:

[0032] If no advertisement selection instruction is received within the preset selection time, the first advertisement will be played on the playback interface.

[0033] Optionally, for each second advertisement, the sum of the duration of the second advertisement and the preset selected duration is less than the duration of the first advertisement.

[0034] Thirdly, embodiments of this application provide an advertising push device applied to a server, the device comprising:

[0035] The advertisement acquisition module is used to acquire multiple advertisements, wherein at least two of the multiple advertisements have different durations, and the different durations represent the different selection rights corresponding to the user choosing different advertisements;

[0036] An advertising sending module is used to send the multiple advertisements to the client in response to the fulfillment of the advertising push conditions, so that the client displays the preview information of the multiple advertisements on the playback interface and plays the advertisement selected by the user based on the preview information, wherein the preview information includes the duration of the advertisement.

[0037] Fourthly, embodiments of this application provide an advertising push device applied to a client, the device comprising:

[0038] An advertisement receiving module is used to receive multiple advertisements sent by a server, wherein at least two of the multiple advertisements have different durations;

[0039] An information display module is used to display preview information of the multiple advertisements on the playback interface, wherein the preview information includes at least the duration of the advertisement, and the duration included in the preview information is used to indicate to the user the different selection benefits corresponding to different advertisements;

[0040] An advertisement playback module is used to display preview information of the multiple advertisements on the playback interface, wherein the preview information includes at least the duration of the advertisement.

[0041] Fifthly, embodiments of this application provide a server, including a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other through the communication bus;

[0042] Memory, used to store computer programs;

[0043] When a processor executes a program stored in memory, it implements any of the steps described in the first aspect.

[0044] Sixthly, embodiments of this application provide a client, including a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other through the communication bus;

[0045] Memory, used to store computer programs;

[0046] When a processor executes a program stored in memory, it implements any of the steps described in the second aspect.

[0047] In a seventh aspect, embodiments of this application provide a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the method described in either the first or second aspect.

[0048] In the solution provided in this application embodiment, the server can acquire multiple advertisements and send them to the client when the advertisement push conditions are met. The client can display preview information of multiple advertisements on the playback interface, and then play the advertisement selected by the user based on the preview information. Since the preview information displayed on the playback interface includes the duration of the advertisement, and different durations correspond to different selection benefits, the user can intuitively understand the duration of each advertisement through the preview information. In order to obtain better selection benefits (e.g., shortening the viewing time of the advertisement), the user can actively select one of the advertisements. By giving users the right to choose and using the different benefits corresponding to different durations as guidance, the user's resistance to advertisements can be reduced, and forced playback can be transformed into active selection, which can improve the user's focus when watching advertisements, thereby improving the conversion rate of advertisements. Of course, implementing any product or method of this application does not necessarily require achieving all of the above-described advantages simultaneously. Attached Figure Description

[0049] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below.

[0050] Figure 1 A flowchart illustrating an advertising push method provided in this application embodiment;

[0051] Figure 2 For based on Figure 1 The illustration shows a first schematic diagram of the preview information display method in the embodiment shown;

[0052] Figure 3 A flowchart illustrating another advertising push method provided in this application embodiment;

[0053] Figure 4 For based on Figure 1 A second schematic diagram illustrating the preview information display method of the illustrated embodiment;

[0054] Figure 5 For based on Figure 1 The diagram illustrates a third method for displaying preview information in the illustrated embodiment.

[0055] Figure 6 For based on Figure 1 The illustrated embodiment shows a fourth schematic diagram of the preview information display method;

[0056] Figure 7 For based on Figure 1 A fifth schematic diagram of the preview information display method in the illustrated embodiment;

[0057] Figure 8 For based on Figure 1A flowchart of an advertising push method according to the embodiment shown;

[0058] Figure 9 This is a schematic diagram of the structure of an advertising push device provided in an embodiment of this application;

[0059] Figure 10 This is a schematic diagram of another advertising push device provided in an embodiment of this application;

[0060] Figure 11 This is a schematic diagram of the structure of a server provided in an embodiment of this application;

[0061] Figure 12 This is a schematic diagram of the structure of a client provided in an embodiment of this application. Detailed Implementation

[0062] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art based on this application are within the scope of protection of this application.

[0063] To improve the conversion rate of advertisements, embodiments of this application provide an advertisement push method, apparatus, server, client, computer-readable storage medium, and computer program product. The advertisement push method provided in this application embodiment will be described first below.

[0064] An ad push method, applied to a server, includes: acquiring multiple ads, wherein at least two of the ads have different durations; in response to meeting ad push conditions, sending the multiple ads to a client, causing the client to display preview information of the multiple ads on a playback interface, and playing an ad selected by the user based on the preview information, wherein the preview information includes the ad duration. Since the preview information displayed on the playback interface includes the ad duration, and different durations correspond to different selection benefits, users can intuitively understand the duration of each ad through the preview information. To obtain better selection benefits (e.g., shortening the ad viewing time), users can actively select one ad. By giving users the right to choose and using the different benefits corresponding to different durations as guidance, user resistance to ads can be reduced, turning forced playback into active selection, increasing user focus while watching ads, thereby improving ad conversion rates.

[0065] like Figure 1 As shown, an advertising push method is applied to a server, the method comprising:

[0066] S101, retrieve multiple ads.

[0067] Among these multiple advertisements, at least two advertisements have different durations, and the different durations represent the different selection rights corresponding to users choosing different advertisements.

[0068] S102, in response to meeting the advertising push conditions, the plurality of advertisements are sent to the client so that the client displays preview information of the plurality of advertisements on the playback interface and plays the advertisement selected by the user based on the preview information.

[0069] The preview information includes the duration of the advertisement.

[0070] As can be seen, in this embodiment, the server can acquire multiple advertisements and send them to the client when the advertisement push conditions are met. The client can display preview information of multiple advertisements on the playback interface, and then play the advertisement selected by the user based on the preview information. Since the preview information displayed on the playback interface includes the duration of the advertisement, and different durations correspond to different selection benefits, the user can intuitively understand the duration of each advertisement through the preview information. In order to obtain better selection benefits (e.g., shortening the viewing time of the advertisement), the user can actively select one of the advertisements. By giving users the right to choose and using the different benefits corresponding to different durations as guidance, the user's resistance to advertisements can be reduced, and forced playback can be transformed into active selection, which can improve the user's focus when watching advertisements, thereby improving the conversion rate of advertisements.

[0071] The aforementioned server can be a backend server for a video playback platform, such as an ad delivery server or a content analysis server, etc., without specific limitations. In step S101, the server can retrieve multiple ads. At least two of the ads must have different durations; that is, the durations of the ads are not entirely the same. For example, assuming the server retrieves three ads, their durations could be 30 seconds, 15 seconds, and 15 seconds respectively.

[0072] In step S102, the server, in response to the fulfillment of the advertising push conditions, can send multiple advertisements to the client. The advertising push conditions indicate the timing of the advertisement push, and can be set according to actual needs, such as before video playback, during video playback, etc., without specific limitations here.

[0073] For example, assuming the ad delivery condition is before video playback, the server can send multiple ads to the client before each video playback. Alternatively, assuming the ad delivery condition is when the video is halfway through playback, the server can send multiple ads to the client each time the video is halfway through playback.

[0074] After receiving multiple advertisements, the client can display preview information of these advertisements on the playback interface. The client can be an application or webpage of a video playback platform, and can run on electronic devices such as mobile phones, tablets, and desktop computers; no specific limitations are made here.

[0075] Preview information can include the duration of the advertisement. For example, assuming an advertisement lasts 15 seconds, the preview information could include "15 seconds," "Ad duration 15 seconds," "Countdown 15 seconds," etc., as long as it reflects the advertisement's duration. See the example below. Figure 2 The playback interface 201 displays Ad A, Ad B and Ad C, with preview information for each ad as follows: ad duration 20 seconds, ad duration 30 seconds and ad duration 15 seconds.

[0076] Since the preview information includes the duration of each ad, users can intuitively understand the duration of multiple ads through the preview information displayed on the playback interface. To shorten the viewing time, users can actively select one ad from multiple ads based on their duration. If a user selects an ad, the client can play that ad on the playback interface. For example, the client can play the ad through the entire playback interface or through a portion of the playback interface; both are reasonable.

[0077] By addressing the dual pain points of user aversion to forced advertising and the desire for shorter ad wait times, a clever game theory mechanism is designed: multiple ads of varying durations are offered for users to choose from. The different ad durations represent different benefits users receive for choosing different ads. Users select an ad based on their preferences, thus actively exchanging their choice for better benefits. This is the first time a user-driven choice mechanism based on duration differences has been introduced into streaming advertising, rather than forcing users to watch an ad. The nature of advertising shifts from passive indoctrination to active transaction, transforming the traditional forced indoctrination model into a time-saving, rational transaction. User aversion to advertising is minimized from the outset, resulting in significantly higher user focus and conversion rates compared to currently used forced advertising, simultaneously improving advertising effectiveness and user experience.

[0078] One possible implementation includes the step of acquiring multiple advertisements, comprising: acquiring a first advertisement selected by an advertising recommendation system according to preset advertising recommendation rules; and acquiring multiple second advertisements generated based on product characteristics and video characteristics of a target video, wherein the target video is a video to be played or currently being played by the client, and the duration of the second advertisement is shorter than that of the first advertisement. On one hand, since the second advertisements are generated based on product characteristics and video characteristics of the target video, and the video characteristics of the target video can reflect the video content of the target video, the product displayed in the second advertisement matches the viewing context of the target video, which can reduce context mismatch caused by relying on historical profiles for ad push. On the other hand, since the duration of the first advertisement is longer than that of the second advertisements, by setting a game structure between the first advertisement and multiple second advertisements, users can be further guided to actively choose advertisements to shorten ad waiting time, thereby reducing user aversion to advertisements.

[0079] The following is a detailed description of this implementation method. The aforementioned plurality of advertisements may include a first advertisement and a plurality of second advertisements, wherein the duration of each second advertisement is shorter than the duration of the first advertisement. For example, assuming the duration of the first advertisement is 30 seconds, the duration of the second advertisements may be 25 seconds, 20 seconds, 15 seconds, and so on.

[0080] The first advertisement can be selected by an advertising recommendation system according to preset advertising recommendation rules. This advertising recommendation system can be the system used by video playback platforms for advertising recommendation in current related technologies; therefore, the first advertisement is also called the default advertisement. Preset advertising recommendation rules can include the user's historical search records on the video playback platform, i.e., historical profiles. Since historical search records can reflect the user's interests, the first advertisement selected according to historical search records can, to some extent, match the user's preferences.

[0081] For example, if a user's search history contains a large number of gaming videos, indicating the user's interest in games, then the first ad could be a gaming ad. Similarly, if a user's search history contains a large number of beauty videos, indicating the user's interest in beauty products, then the first ad could be a beauty ad.

[0082] The second advertisement can be generated based on product characteristics and the video characteristics of the target video. Compared to the first advertisement, the second advertisement is an additional optional feature, hence it is also called an optional advertisement. The target video can be a video that the client is about to play or is currently playing. Specifically, if the advertisement is pushed before video playback, the target video can be the video that the client is about to play; if the advertisement is pushed during video playback, the target video can be the video that the client is currently playing.

[0083] The video features of the target video can be its content-level characteristics, such as the context, type, and values ​​it conveys. Product features can be the characteristics of the product showcased in the second advertisement, such as the product type and target audience.

[0084] Building on this, since the second ad is generated based on product characteristics and the video characteristics of the target video, the contextual matching between the product displayed in the second ad and the target video can reduce contextual mismatches caused by relying on historical profiles for ad delivery. For example, assuming the target video is a serious documentary, even if the user has watched comedy games multiple times before, the product displayed in the second ad will not be a comedy game because the viewing context of comedy games does not match that of the documentary.

[0085] One possible implementation method for generating the second advertisement includes: acquiring video features of the target video, wherein the video features include character features, which are obtained by extracting character profiles of characters in the target video; determining target products to be advertised when playing the target video from among products that need to be advertised, based on the degree of matching between product features and video features; generating advertising copy for each target product based on the product features of each target product and the video features; and generating a second advertisement corresponding to each target product based on the advertising copy of each target product.

[0086] Currently, while contextual advertising is related to video content to some extent (e.g., playing car ads within car programs), it lacks novelty and targeting. The solution provided in this application, however, determines the copy of the second advertisement based on product and video characteristics. Since video characteristics can include character and contextual features, the second advertisement is a dynamically generated role-based contextual ad. The product displayed is strongly correlated with the key character, possessing strong immediate relevance, appeal, and entertainment value. It fully utilizes the most attractive element in film and television works—the characters. Thus, the advertisement is no longer an intrusion but a derivative or Easter egg of the video, possessing a strong sense of immersion and nativeness. This fundamentally solves the problem of why users actively choose advertisements, while ensuring that when users actively choose, they tend to select advertisements they are genuinely interested in, achieving ultra-precise contextual marketing.

[0087] The following is a detailed description of this implementation method. The video features of the target video can be extracted in real time during the playback of the target video, or they can be extracted in advance before the playback of the target video; no specific limitation is made here. The video features of the target video can be obtained by a multimodal AI (Artificial Intelligence) model performing in-depth analysis of the content of the target video.

[0088] Specifically, multimodal AI models can extract contextual features from target videos through computer vision and natural language processing (NLP). They can also identify key characters in target videos, extract character personas (character features), and then use contextual features and character features as video features of the target video.

[0089] Contextual features can reflect the type of video, such as: documentary, science fiction, art film, etc. When the target video is a TV series or movie, key characters can include the protagonist and specific supporting characters. The protagonist can be identified through the cast and crew list, while specific supporting characters can be determined based on the popularity of the actor playing the role and the level of discussion surrounding the character—that is, the actor playing the specific supporting role is highly popular, or the specific supporting character is widely discussed in online communities.

[0090] Key characters' traits can be categorized into explicit and implicit traits. Explicit traits are those conveyed visually, such as items the character uses (e.g., cell phone, watch, car), clothing style, and profession (e.g., lawyer, doctor, agent). Implicit traits are those conveyed through the plot, such as personality traits (e.g., bravery, intelligence, humor, fashion sense), current emotions, and values ​​or needs revealed in dialogue.

[0091] To maximize the likelihood that users watching the target video are the audience for the product being advertised, after obtaining the video features of the target video, the server can retrieve the target product for which the advertisement will be pushed from the advertising creative library based on the degree of matching between the video features and product features. The advertising creative library can include all the products for which advertisements will be pushed.

[0092] It should be noted that the above content is based on the premise that video features include both character features and context features. In addition, the server may also extract only context features or character features as video features, and this application embodiment does not specifically limit this.

[0093] Next, the server can generate advertising copy for each target product based on its product characteristics and the video characteristics of the target video. Specifically, the server can input the product characteristics and video characteristics into a Large Language Model (LLM) and obtain the advertising copy generated by the LLM combining the product characteristics and video characteristics.

[0094] For example, suppose character A in a movie or TV show is characterized as: agent, high-tech, adventurous, and uses a laptop from brand A. Then, brand A laptops and brand B coffee from the advertising creative library can be used as target products. The server can input the aforementioned character characteristics and the name of brand A laptops into a large language model, which can then generate the advertising copy: "Complete impossible missions like an agent. Brand A laptops: ultimate performance, the agent's choice." Similarly, the server can input the aforementioned character characteristics and brand B coffee into the large language model, which can then generate the advertising copy: "Want to have the composure and calmness of an agent? Try brand B coffee, get into the zone in a second."

[0095] After generating the advertising copy for each target product, the server can, based on the advertising copy, generate at least one second advertisement for each target product using a generative artificial intelligence (AI) model. The generative AI model can be a text-to-image model or a text-to-video model, such as an AIGC (Artificial Intelligence Generated Content) model.

[0096] As AIGC technology matures, it has demonstrated high fidelity in generating creative content with extremely low marginal costs. Therefore, large-scale, in-depth, and personalized ad generation using AIGC technology can create highly relevant creative ads for a vast number of film and television works at low cost, making it economically feasible like never before.

[0097] Of course, the method of generating the first advertisement can be the same as or different from that of generating the second advertisement. For example, the first advertisement can be obtained by shooting real-life scenes, which is also reasonable. Therefore, we will not limit the method of generating the first advertisement here.

[0098] One possible implementation involves determining the target product for ad push when playing the target video, based on the matching degree between product features and video features. This step includes: for each product requiring ad push, determining a first semantic correlation between the product features and the character features, and a second semantic correlation between the product features and the contextual features; identifying products whose first semantic correlation is greater than a first preset threshold and whose second semantic correlation is greater than a second preset threshold as target products for ad push when playing the target video. Since both the first and second semantic correlations corresponding to the target products are relatively high, and the first semantic correlation reflects the correlation between the target product's product features and the character features, while the second semantic correlation reflects the correlation between the target product's product features and the contextual features, the target products are deeply associated with the character features and contextual features of the target video. This results in higher user acceptance of the target products, improving ad conversion rates.

[0099] For each product that needs to be advertised, i.e., the product that advertises to the video playback platform, the server can determine the first semantic relevance between the product features and the role features, and determine the second semantic relevance between the product features and the context features.

[0100] Specifically, the server can convert product features, role features, and context features into word vectors, determine the similarity between the word vectors of product features and the word vectors of role features as the first semantic relevance, and determine the similarity between the word vectors of product features and the word vectors of context features as the second semantic relevance.

[0101] If a product's features have both high primary and secondary semantic relevance, it indicates a high degree of association between the product's features and the role features, as well as a high degree of association with the contextual features. In this case, pushing an ad for the product while playing the target video can prevent users from being distracted and reduce user aversion.

[0102] To identify products that meet the above conditions, the server can determine products whose first semantic relevance is greater than a first preset threshold and whose second semantic relevance is greater than a second preset threshold as target products. The first and second preset thresholds can be set according to actual needs, such as 80%, 90%, 95%, etc., and are not specifically limited here.

[0103] As shown in Table 1 below, assuming the first preset threshold is 90% and the second preset threshold is 85%, then among the following products, only the desk lamp has a first semantic relevance greater than the first preset threshold and a second semantic relevance greater than the second preset threshold. Therefore, the desk lamp can be used as the target product for advertising push when playing the campus drama.

[0104] Table 1. Semantic Relevance of Product Features with Role Features and Contextual Features

[0105] One possible implementation includes the step of acquiring multiple advertisements, comprising: acquiring a user profile of a user using the client, wherein the user profile reflects the user's preference for advertisements; acquiring multiple advertisements based on the user profile; and after the step of sending the multiple advertisements to the client, the method further comprises: determining the advertisement selected by the user based on the preview information; updating the user profile based on information of the advertisement selected by the user, information of the advertisement not selected by the user among the multiple advertisements, and video features of a target video, wherein the target video is: a video to be played or currently being played by the client.

[0106] Current technologies do not display multiple ads on the playback interface; the data collected based on user actions during ad playback is limited to "exposure" or "skipping." However, the solution provided in this application displays multiple ads simultaneously for the user to choose from, and the user's ad selection behavior itself is a highly valuable data feedback. Therefore, based on the user's selection behavior, it can be collected that the user actively selected a specific ad while watching a video, rather than the others. This allows for the collection of high-dimensional, high-intent, and high-precision real-time data points, which can then be used to update user profiles, providing a strong basis for subsequent ad and video recommendations.

[0107] The following is a detailed description of this implementation method. When the server receives multiple advertisements to be pushed to a client, it can obtain a user profile of the user using that client, and then obtain multiple advertisements based on the obtained user profile. The aforementioned user profile can be a text tag describing the user's preferences for advertisements, such as: "interested in high-tech products", "prefers beauty products", "likes actor so-and-so", etc.

[0108] In this scenario, when retrieving multiple ads, the server can select the ads with the highest third semantic relevance based on the third semantic relevance between the user profile and the ad's features. These ad features may include: product characteristics of the displayed product, features of the ad copy, the spokesperson, the overall atmosphere of the ad, etc., without specific limitations.

[0109] As shown in Table 2 below, assuming that the third semantic relevance between the features of Ads 1-5 and the user profile is 91%, 43%, 89%, 31%, and 96% respectively, and the server needs to obtain 3 ads, then the server can obtain Ads 1, Ads 3, and Ads 5.

[0110] Table 2. Characteristics of Advertisements and User Profiles

[0111] After retrieving multiple advertisements based on the user profile, the server can send these advertisements to the client. The client can then display previews of these advertisements on the playback screen. If the user actively selects an advertisement on the playback screen, the server can determine the selected advertisement and update the user's profile based on the information of the selected advertisement, the information of the advertisements the user did not select, and the video features of the target video.

[0112] Specifically, when updating user profiles, if the existing text tags do not contain relevant content, the server can add text tags, such as adding the text tag "interested in high-tech products". If the existing text tags contain relevant content, the server can enhance the existing text tags, such as enhancing the existing text tag "attracted by professional advertising copy" to "attracted by professional / efficient advertising copy".

[0113] For example, if a user is watching a spy video, the multimodal AI model identifies the video's features as follows: contextual features—"spy / agent theme," and character features—"agent character (using high-tech equipment)." The playback interface displays Ad A, Ad B, and Ad C. Ad A is a default ad selected by the ad recommendation system that is unrelated to the spy video; Ad B is a computer ad shown by the agent in the spy video; and Ad C is a coffee ad shown by the agent in the spy video.

[0114] Suppose a user actively selects ad B but not ad C. The server captures this selection behavior and associates the information of the "ad selected by the user" (ad B: computer, high-tech product), the information of the "ad not selected by the user" (ad C: coffee, beverage), and the video characteristics of the current target video (espionage scenario, agent role), comprehensively updating the user profile. After the update, the following high-dimensional composite tags can be formed: {"Current Context Preference: Espionage / Agent Theme", "Role Association Preference: Agent Role + High-Tech Equipment", "Product Preference: High-Tech Product", "Ad Copy Preference: Attracted by 'Professional / Efficient' Style", "Selection Behavior Characteristics: Preference for Actively Selecting Ads"}.

[0115] For example, if a user always chooses the default ad, the user profile can be updated by adding the text tags "insensitive to ads" or "not interested in any of the currently available ads." Since the user is insensitive to ads, the number of available ads can be reduced when pushing ads in the future, thus saving computing power.

[0116] In accordance with the above-mentioned advertising push method, this application embodiment also provides another advertising push method, which will be described below.

[0117] like Figure 3 As shown, an advertising push method is applied to a client, the method comprising:

[0118] S301 receives multiple advertisements sent by the server.

[0119] Among the multiple advertisements, at least two advertisements have different durations.

[0120] S302, display preview information of the multiple advertisements on the playback interface.

[0121] The preview information includes at least the duration of the advertisement, and the duration included in the preview information is used to indicate to the user the different selection benefits corresponding to different advertisements.

[0122] S303, in response to the user's ad selection instruction based on the preview information, the ad indicated by the ad selection instruction is played on the playback interface.

[0123] As can be seen, in this embodiment, the client can receive multiple advertisements sent by the server and display preview information of these advertisements on the playback interface. The client then plays the advertisement selected by the user based on the preview information. Since the preview information displayed on the playback interface includes the duration of the advertisement, and different durations correspond to different selection benefits, the user can intuitively understand the duration of each advertisement through the preview information. To obtain better selection benefits (e.g., shortening the viewing time of the advertisement), the user can actively select one of the advertisements. By giving users the right to choose and using the different benefits corresponding to different durations as guidance, the user's resistance to advertisements can be reduced, turning forced playback into active selection, which can improve the user's focus when watching advertisements, thereby increasing the conversion rate of advertisements.

[0124] In step S301, the client can obtain multiple advertisements sent by the server. Next, the client can display preview information of these advertisements on the playback interface, i.e., proceed to step S302. The display method of the preview information can be determined based on the type of electronic device the client is using.

[0125] For example, when the client is running on a landscape device such as a tablet or desktop computer, the preview information can be displayed horizontally, such as... Figure 2 As shown. When the client is running on a portrait-oriented device such as a mobile phone, the preview information can be displayed vertically, as shown below. Figure 4 As shown, the preview information may include: Ad A has an ad duration of 20 seconds, Ad B has an ad duration of 30 seconds, and Ad C has an ad duration of 15 seconds.

[0126] In response to an ad selection command issued by the user based on preview information, the client can play the ad specified in the command on the playback interface. The ad selection command can be triggered by a single click, double click, long press, etc., without specific limitations here.

[0127] It should be noted that since multiple advertisements have been pre-delivered to the client by the server in step S301, the client can cache them locally after receiving the multiple advertisements sent by the server. In this way, when the user issues an advertisement selection command, the client can directly retrieve the corresponding advertisement from the local cache for playback without sending another advertisement retrieval request to the server. This ensures that the "time-saving" benefit obtained by the user in actively selecting short-duration advertisements can be truly realized, avoiding the network transmission delay caused by real-time request for advertisements from the server, which would instead consume more time and affect the user experience.

[0128] In one possible implementation, the plurality of advertisements includes a first advertisement and a plurality of second advertisements, wherein the duration of the second advertisements is shorter than the duration of the first advertisement; after the step of displaying preview information of the plurality of advertisements on the playback interface, the method further includes: if no advertisement selection instruction is received within a preset selection duration, playing the first advertisement on the playback interface. Thus, even if the user does not issue an advertisement selection instruction within the preset selection duration, the client can still successfully complete the advertisement playback.

[0129] The preview information for the aforementioned advertisements may also include at least one of the following: a preview image of the advertisement, the name of the product displayed in the advertisement, and the spokesperson for the advertisement. The preview image of the advertisement may be a frame from the advertisement.

[0130] The following examples, with reference to the attached diagram, illustrate how the preview information is displayed:

[0131] Assuming multiple advertisements include Ad 1, Ad 21, and Ad 22, then see... Figure 5 The preview information for the second advertisement 1 and the second advertisement 2 can be displayed on either side of the preview information for the first advertisement; see also Figure 6The preview information for the first advertisement can be displayed on the left side of the playback interface, and the preview information for the second advertisement (ad 1) and the second advertisement (ad 2) can be displayed on the right side of the playback interface; see also Figure 7 The preview information of the first advertisement can be displayed on the entire playback interface, and the preview information of the second advertisement 1 and the second advertisement 2 can be displayed on the preview information of the first advertisement. This is all reasonable, and there is no limitation on the display method of the preview information.

[0132] After displaying previews of multiple advertisements on the playback interface, if the client does not receive an advertisement selection instruction from the user within the preset selection duration, the client can play the first advertisement on the playback interface. The preset selection duration can be set according to actual needs, such as 3 seconds, 5 seconds, 7 seconds, etc., and is not limited here.

[0133] Since users also need to spend some time actively selecting ads, in order to incentivize users to actively select ads, the duration of the second ad can be further shortened. Specifically, for each second ad, the sum of the duration of the second ad and the preset selection duration is less than the duration of the first ad.

[0134] For example, assuming the preset selection duration is 5 seconds and the first advertisement duration is 30 seconds, then the second advertisement duration will not exceed 25 seconds.

[0135] In one possible implementation, if a user issues an ad selection command for either the first or second ad, but skips the second ad during its playback, the client can display feedback options on the playback interface for the user to choose from. These feedback options would list the reasons why the user did not select the second ad or skipped it.

[0136] For example, feedback options could be: "Why didn't you choose the second ad? A. Not interested in the product; B. Not interested in the spokesperson; C. Not interested in the creative style."

[0137] In this way, after a user selects an option, the server can use the displayed second advertisement as a negative sample to further guide the creative generation direction of the second advertisement, thus realizing a negative feedback mechanism.

[0138] One possible implementation is that after a user selects a second advertisement, the server can anonymize the composite data of "viewing context, character, and user selection" and provide it as a high-value data asset to the product team in real time. For example, the server could inform the product team: "When watching the 'intelligent character' storyline, the conversion rate of users who selected your product increased by 20%."

[0139] One possible implementation, such as Figure 8 As shown, advertising delivery methods can include:

[0140] S801, Analyze Characters and Storyline. The server can analyze the character and situational characteristics of a target video before playback, using these as video features.

[0141] S802, Ad Delivery. The server can record the products delivered by the product provider and analyze their characteristics.

[0142] S803, generate a second advertisement. The server can generate multiple second advertisements based on product characteristics and video characteristics.

[0143] S804, Approved. The server can review the second advertisement at the risk control level to determine whether it violates the Advertising Law. If it does not violate the Advertising Law, the review is approved.

[0144] S805, Selecting a first advertisement and multiple second advertisements. While a user is watching a target video through the client, the server can use an ad recommendation algorithm to select a first advertisement and multiple second advertisements, and send the selected advertisements to the client.

[0145] S806, Detecting Ad Selection Commands. The client can display preview information of the first ad and multiple second ads on the playback interface, along with prompts to guide the user to actively select an ad, such as: "Are you willing to sacrifice 5 seconds of clicking to save 15 seconds of ad time?" If an ad selection command is received from the user, the client can play the ad indicated by the command.

[0146] It's important to note that the prompts and preview information work together within the interface to guide users in making choices. Specifically, the preview information presents the user with objective data showing the duration of each ad, allowing them to clearly compare the duration differences between ads (e.g., the first ad is "30 seconds" long, and each of the second ads is "15 seconds" long). The prompts, on the other hand, provide a guiding interpretation of these duration differences. They transform the objective duration figures into a relationship between "active choice" and "time gain" that users can intuitively perceive, informing them in simple terms: "You only need to spend a few seconds making a choice to gain the benefit of significantly reduced ad viewing time."

[0147] In other words, the preview information shows users which ads are available and what benefits each ad offers, while the prompts explain to users the actual benefits they can gain from making a choice. Together, they guide users to make active choices.

[0148] S807, Update User Profile. The server can determine the advertisement selected by the user and update the user profile based on the information of that advertisement to guide the advertisement recommendation algorithm in subsequent advertisement recommendations.

[0149] Corresponding to the above-described advertising push method, this application also provides an advertising push device. The advertising push device provided in this application embodiment is described below.

[0150] like Figure 9 As shown, an advertising push device is applied to a server, the device comprising:

[0151] The advertisement acquisition module 901 is used to acquire multiple advertisements, wherein at least two of the multiple advertisements have different durations, and the different durations represent the different selection rights corresponding to the user choosing different advertisements;

[0152] The advertisement sending module 902 is used to send the plurality of advertisements to the client in response to the fulfillment of the advertisement push conditions, so that the client displays the preview information of the plurality of advertisements on the playback interface and plays the advertisement selected by the user based on the preview information, wherein the preview information includes the duration of the advertisement.

[0153] As can be seen, in this embodiment, the server can acquire multiple advertisements and send them to the client when the advertisement push conditions are met. The client can display preview information of multiple advertisements on the playback interface, and then play the advertisement selected by the user based on the preview information. Since the preview information displayed on the playback interface includes the duration of the advertisement, and different durations correspond to different selection benefits, the user can intuitively understand the duration of each advertisement through the preview information. In order to obtain better selection benefits (e.g., shortening the viewing time of the advertisement), the user can actively select one of the advertisements. By giving users the right to choose and using the different benefits corresponding to different durations as guidance, the user's resistance to advertisements can be reduced, and forced playback can be transformed into active selection, which can improve the user's focus when watching advertisements, thereby improving the conversion rate of advertisements.

[0154] In one possible implementation, the advertisement acquisition module 901 may include:

[0155] The first advertisement acquisition submodule is used to acquire the first advertisement selected by the advertisement recommendation system according to the preset advertisement recommendation rules.

[0156] The second advertisement acquisition submodule is used to acquire multiple second advertisements generated based on product characteristics and the video characteristics of the target video, wherein the target video is the video to be played or currently being played by the client, and the duration of the second advertisement is shorter than the duration of the first advertisement.

[0157] In one possible implementation, the above-described apparatus may further include:

[0158] The feature acquisition module is used to acquire video features of the target video, wherein the video features include character features, which are obtained by extracting character portraits of the characters in the target video;

[0159] The target product determination module is used to determine the target product to be pushed with ads when playing the target video from the products that need to be pushed with ads, based on the degree of matching between the product characteristics and the video characteristics.

[0160] The advertising copy generation module is used to generate advertising copy for each target product based on the product characteristics of each target product and the video characteristics.

[0161] The second ad generation module is used to generate a second ad for each target product based on the ad copy for each target product.

[0162] In one possible implementation, the video features include character features and contextual features. In this case, the target product determination module may include:

[0163] The first relevance determination submodule is used to determine the first semantic relevance between the product features and the role features of each product that needs to be pushed with ads, and to determine the second semantic relevance between the product features and the context features.

[0164] The second relevance determination submodule is used to determine products whose corresponding first semantic relevance is greater than a first preset threshold and whose corresponding second semantic relevance is greater than a second preset threshold, and to use these products as target products for advertising push when playing the target video.

[0165] In one possible implementation, the advertisement acquisition module 901 may include:

[0166] The user profile acquisition submodule is used to acquire user profiles of users using the client, wherein the user profiles reflect the user's preferences for advertisements;

[0167] The advertisement acquisition submodule is used to acquire multiple advertisements based on the user profile;

[0168] In this case, the above-mentioned device may further include:

[0169] An advertisement determination module is used to determine the advertisement selected by the user based on the preview information;

[0170] The user profile update module is used to update the user profile based on the information of the advertisement selected by the user, the information of the advertisement not selected by the user among the multiple advertisements, and the video features of the target video, wherein the target video is: the video to be played or currently being played by the client.

[0171] Corresponding to the other advertising push method described above, this application embodiment also provides another advertising push device. The following describes the other advertising push device provided in this application embodiment.

[0172] like Figure 10 As shown, an advertising push device is applied to a client, the device comprising:

[0173] The advertisement receiving module 1001 is used to receive multiple advertisements sent by the server, wherein at least two of the multiple advertisements have different durations;

[0174] The information display module 1002 is used to display preview information of the multiple advertisements on the playback interface, wherein the preview information includes at least the duration of the advertisement, and the duration included in the preview information is used to indicate to the user the different selection benefits corresponding to different advertisements.

[0175] The advertisement playback module 1003 is used to display preview information of the plurality of advertisements on the playback interface, wherein the preview information includes at least the duration of the advertisement.

[0176] As can be seen, in this embodiment, the client can receive multiple advertisements sent by the server and display preview information of these advertisements on the playback interface. The client then plays the advertisement selected by the user based on the preview information. Since the preview information displayed on the playback interface includes the duration of the advertisement, and different durations correspond to different selection benefits, the user can intuitively understand the duration of each advertisement through the preview information. To obtain better selection benefits (e.g., shortening the viewing time of the advertisement), the user can actively select one of the advertisements. By giving users the right to choose and using the different benefits corresponding to different durations as guidance, the user's resistance to advertisements can be reduced, turning forced playback into active selection, which can improve the user's focus when watching advertisements, thereby increasing the conversion rate of advertisements.

[0177] In one possible implementation, the plurality of advertisements includes a first advertisement and a plurality of second advertisements, wherein the duration of the second advertisements is shorter than the duration of the first advertisements. In this case, the device may further include:

[0178] The first advertisement playback module is used to play the first advertisement on the playback interface if no advertisement selection instruction is received within a preset selection time.

[0179] One possible implementation is that, for each second advertisement, the sum of the duration of the second advertisement and the preset selection duration is less than the duration of the first advertisement.

[0180] This application also provides a server, such as... Figure 11 As shown, it includes:

[0181] Memory 1101 is used to store computer programs;

[0182] The processor 1102 is used to execute the program stored in the memory 1101 to implement an advertising push method as described in any of the above embodiments.

[0183] Furthermore, the aforementioned server may also include a communication bus and / or a communication interface, and the processor 1102, the communication interface, and the memory 1101 communicate with each other through the communication bus.

[0184] As can be seen, in this embodiment, the server can acquire multiple advertisements and send them to the client when the advertisement push conditions are met. The client can display preview information of multiple advertisements on the playback interface, and then play the advertisement selected by the user based on the preview information. Since the preview information displayed on the playback interface includes the duration of the advertisement, and different durations correspond to different selection benefits, the user can intuitively understand the duration of each advertisement through the preview information. In order to obtain better selection benefits (e.g., shortening the viewing time of the advertisement), the user can actively select one of the advertisements. By giving users the right to choose and using the different benefits corresponding to different durations as guidance, the user's resistance to advertisements can be reduced, and forced playback can be transformed into active selection, which can improve the user's focus when watching advertisements, thereby improving the conversion rate of advertisements.

[0185] This application also provides a client, such as... Figure 12 As shown, it includes:

[0186] Memory 1201 is used to store computer programs;

[0187] When the processor 1202 executes the program stored in the memory 1201, it implements another advertising push method as described in any of the above embodiments.

[0188] Furthermore, the aforementioned client may also include a communication bus and / or a communication interface, and the processor 1202, the communication interface, and the memory 1201 communicate with each other through the communication bus.

[0189] As can be seen, in this embodiment, the client can receive multiple advertisements sent by the server and display preview information of these advertisements on the playback interface. The client then plays the advertisement selected by the user based on the preview information. Since the preview information displayed on the playback interface includes the duration of the advertisement, and different durations correspond to different selection benefits, the user can intuitively understand the duration of each advertisement through the preview information. To obtain better selection benefits (e.g., shortening the viewing time of the advertisement), the user can actively select one of the advertisements. By giving users the right to choose and using the different benefits corresponding to different durations as guidance, the user's resistance to advertisements can be reduced, turning forced playback into active selection, which can improve the user's focus when watching advertisements, thereby increasing the conversion rate of advertisements.

[0190] The communication bus mentioned above for the server and client can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. This communication bus can be divided into address bus, data bus, control bus, etc. For ease of illustration, only one thick line is used to represent it in the diagram, but this does not mean that there is only one bus or one type of bus. The communication interface is used for communication between the aforementioned terminal and other devices.

[0191] The memory may include random access memory (RAM) or non-volatile memory, such as at least one disk storage device. Optionally, the memory may also be at least one storage device located remotely from the aforementioned processor.

[0192] The processors mentioned above can be general-purpose processors, including central processing units (CPUs), network processors (NPs), etc.; they can also be digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.

[0193] In another embodiment provided in this application, a computer-readable storage medium is also provided, wherein a computer program is stored therein, and when the computer program is executed by a processor, it implements any of the advertising push methods described in the above embodiments.

[0194] In another embodiment provided in this application, a computer program product containing instructions is also provided, which, when run on a computer, causes the computer to execute any of the advertising push methods described in the above embodiments.

[0195] In the above embodiments, implementation can be achieved entirely or partially through software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented entirely or partially as a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that integrates one or more available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., solid state disk (SSD)).

[0196] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0197] The various embodiments in this specification are described in a related manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the embodiments for apparatus, server, client, computer-readable storage medium, and computer program products are basically similar to the method embodiments, and therefore the descriptions are relatively simple; relevant parts can be referred to the descriptions of the method embodiments.

[0198] The above description is merely a preferred embodiment of this application and is not intended to limit the scope of protection of this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application are included within the scope of protection of this application.

Claims

1. An advertising push method, characterized in that, Applied to a server, the method includes: Multiple advertisements are obtained, wherein at least two of the advertisements have different durations, and the different durations represent the different selection rights corresponding to the user choosing different advertisements; In response to the fulfillment of the advertising push conditions, the plurality of advertisements are sent to the client so that the client displays preview information of the plurality of advertisements on the playback interface and plays the advertisement selected by the user based on the preview information, wherein the preview information includes the duration of the advertisement.

2. The method according to claim 1, characterized in that, The step of obtaining multiple advertisements includes: Obtain the first advertisement selected by the advertising recommendation system according to preset advertising recommendation rules; Multiple second advertisements generated based on product characteristics and the video characteristics of the target video are obtained, wherein the target video is the video to be played or currently being played by the client, and the duration of the second advertisement is shorter than the duration of the first advertisement.

3. The method according to claim 2, characterized in that, The second advertisement is generated in the following ways: The video features of the target video are obtained, wherein the video features include character features, which are obtained by extracting character profiles of the characters in the target video; Based on the degree of matching between product features and video features, the target product to be pushed with ads when playing the target video is determined from the products that need to be pushed with ads. Based on the product characteristics of each target product and the video characteristics, generate advertising copy for each target product; Based on the advertising copy for each target product, generate a second advertisement corresponding to each target product.

4. The method according to claim 3, characterized in that, The video features include character features and contextual features; The step of determining the target product to be pushed an advertisement to when playing the target video from the products that need to be pushed an advertisement to, based on the degree of matching between the product characteristics and the video characteristics, includes: For each product that needs to be pushed with advertisements, determine the first semantic relevance between the product features and the role features, and determine the second semantic relevance between the product features and the context features; Products whose first semantic relevance is greater than a first preset threshold and whose second semantic relevance is greater than a second preset threshold are identified as target products for advertising push when playing the target video.

5. The method according to claim 1, characterized in that, The step of obtaining multiple advertisements includes: Obtain user profiles of users using the client, wherein the user profiles reflect users' preferences for advertisements; Based on the user profile, multiple advertisements are obtained; After the step of sending the plurality of advertisements to the client, the method further includes: Determine the advertisement selected by the user based on the preview information; The user profile is updated based on the information of the advertisement selected by the user, the information of the advertisements not selected by the user among the multiple advertisements, and the video features of the target video, wherein the target video is: the video to be played or currently being played by the client.

6. An advertising push method, characterized in that, Applied to a client, the method includes: Receive multiple advertisements sent by the server, wherein at least two of the multiple advertisements have different durations; The playback interface displays preview information of the multiple advertisements, wherein the preview information includes at least the duration of the advertisement, and the duration included in the preview information is used to indicate to the user the different selection benefits corresponding to different advertisements; In response to an ad selection instruction issued by the user based on the preview information, the ad indicated by the ad selection instruction is played on the playback interface.

7. The method according to claim 6, characterized in that, The plurality of advertisements includes a first advertisement and a plurality of second advertisements, wherein the duration of the second advertisements is shorter than the duration of the first advertisements; After the step of displaying preview information of the multiple advertisements on the playback interface, the method further includes: If no advertisement selection instruction is received within the preset selection time, the first advertisement will be played on the playback interface.

8. The method according to claim 7, characterized in that, For each second advertisement, the sum of the duration of the second advertisement and the preset selection duration is less than the duration of the first advertisement.

9. An advertising push device, characterized in that, Applied to a server, the device includes: The advertisement acquisition module is used to acquire multiple advertisements, wherein at least two of the multiple advertisements have different durations, and the different durations represent the different selection rights corresponding to the user choosing different advertisements; An advertising sending module is used to send the multiple advertisements to the client in response to the fulfillment of the advertising push conditions, so that the client displays the preview information of the multiple advertisements on the playback interface and plays the advertisement selected by the user based on the preview information, wherein the preview information includes the duration of the advertisement.

10. An advertising push device, characterized in that, Applied to a client, the device includes: An advertisement receiving module is used to receive multiple advertisements sent by a server, wherein at least two of the multiple advertisements have different durations; An information display module is used to display preview information of the multiple advertisements on the playback interface, wherein the preview information includes at least the duration of the advertisement, and the duration included in the preview information is used to indicate to the user the different selection benefits corresponding to different advertisements; An advertisement playback module is used to display preview information of the multiple advertisements on the playback interface, wherein the preview information includes at least the duration of the advertisement.

11. A server, characterized in that, It includes a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other through the communication bus; Memory, used to store computer programs; A processor, when executing a program stored in memory, implements the method described in any one of claims 1-5.

12. A client, characterized in that, It includes a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other through the communication bus; Memory, used to store computer programs; A processor, when executing a program stored in memory, implements the method of any one of claims 6-8.

13. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the method described in any one of claims 1-5 or 6-8.

Images