An advertisement putting optimization method, device and equipment based on artificial intelligence cooperation and medium

Abstract

The application provides an advertisement putting optimization method, device and equipment based on artificial intelligence cooperation and a medium, and belongs to the technical field of data processing. The method comprises the following steps: in response to receiving an advertisement putting task, obtaining advertisement content of an advertisement to be put, and determining an advertisement type of the advertisement to be put according to the advertisement content; obtaining an advertisement bid of the advertisement to be put, and determining a plurality of actual playing platforms according to the advertisement bid; obtaining video categories, playing time lengths and playing time periods of the actual playing platforms, and matching the advertisement to be put with the actual playing platforms to obtain a target playing platform based on the advertisement type and the playing time lengths of the video categories; determining a putting strategy of the advertisement to be put according to the playing time periods of the video categories; and putting the advertisement to be put on the target playing platform based on the putting strategy. The application can effectively optimize the advertisement putting effect.

Description

Technical Field

[0001] This application belongs to the field of data processing technology, and in particular relates to a method, apparatus, device and medium for optimizing advertising delivery based on artificial intelligence collaboration. Background Technology

[0002] In today's rapidly developing digital economy, advertising has become a core tool for businesses to enhance brand awareness, boost product sales, and expand market share. Whether it's online e-commerce platforms, social media, video websites, or offline smart screens and outdoor advertising spaces, advertising scenarios are becoming increasingly diverse, and audience reach paths are becoming more complex. For advertisers, the core demands of advertising always revolve around accurately targeting the right audience and maximizing return on investment—ensuring that advertising content reaches genuinely interested users with the means to spend money, avoiding wasting resources on ineffective audiences, while maximizing ad exposure within a limited budget. Against this backdrop, optimizing advertising campaigns is particularly crucial.

[0003] In related technologies, advertisements are directly inserted when the video is paused, started, or ended. However, direct insertion of advertisements may result in a weak correlation between the advertisement and the video content. When the correlation between the advertisement and the video content is weak, users may close the advertisement directly, resulting in a small audience reach after the advertisement is delivered. Therefore, it can be seen that the related technologies are not very effective for advertising. Summary of the Invention

[0004] To address the aforementioned technical issues, this application provides a method, apparatus, device, and medium for optimizing advertising delivery based on artificial intelligence collaboration.

[0005] Firstly, this application provides an advertising delivery optimization method based on artificial intelligence collaboration, employing the following technical solution:

[0006] An AI-based collaborative method for optimizing ad delivery includes:

[0007] In response to receiving an advertising delivery task, the system obtains the advertising content of the advertisement to be delivered and determines the advertising type of the advertisement to be delivered based on the advertising content.

[0008] Obtain the advertising quote for the advertisement to be placed, and determine several actual playback platforms based on the advertising quote;

[0009] The video types, playback durations, and playback periods of each actual playback platform are obtained, and based on the ad type and the playback duration of each video type, the ad to be delivered is matched with each actual playback platform to obtain the target playback platform.

[0010] The placement strategy for the advertisements to be placed is determined based on the playback time of each video type.

[0011] The advertisement to be advertised will be advertised on the target playback platform based on the aforementioned advertising strategy.

[0012] In one embodiment, the advertising content includes advertising images and advertising copy, and determining the advertising type of the advertisement to be delivered based on the advertising content includes:

[0013] The advertisement image is input into a pre-trained advertisement recognition model to obtain the first advertisement type of the advertisement to be delivered. The first advertisement type is used to describe the application scenario of the advertisement to be delivered.

[0014] Obtain several keywords corresponding to multiple preset second ad types, wherein the preset second ad types are used to describe the attributes of the ad to be placed;

[0015] The keywords are matched with the advertising copy to obtain the target keywords for the advertisement to be placed.

[0016] Determine the keyword frequency of each target keyword, and determine the target second ad type corresponding to the ad to be delivered based on the keyword frequency of each target keyword.

[0017] In one embodiment, the step of matching the advertisement to be delivered with each of the actual playback platforms to obtain the target playback platform based on the advertisement type and the playback duration of each of the video types includes:

[0018] Based on the playback duration of each video type, determine the average playback duration of each actual playback platform;

[0019] The playback duration of the advertisement to be delivered is obtained, and the playback duration of the advertisement to be delivered is compared with the average playback duration to determine the initial playback platform, wherein the average playback duration of the initial playback platform is greater than the playback duration of the advertisement to be delivered.

[0020] Determine the first user profile for each of the aforementioned video types;

[0021] The playback frequency of each video type is obtained, and the second user profile of each initial playback platform is determined based on the playback frequency and the first user profile corresponding to each video type.

[0022] The compatibility between the advertisement to be delivered and each of the initial playback platforms is determined based on the advertisement type and the second user profile of each initial playback platform.

[0023] The target playback platform is determined based on the compatibility.

[0024] In one embodiment, determining the second user profile of each initial playback platform based on the first user profile corresponding to each playback frequency and each video type includes:

[0025] For each initial playback platform, obtain the user stickiness rate corresponding to each video type;

[0026] Based on the correspondence between the playback frequency, the playback frequency and the correction coefficient, determine the correction coefficient corresponding to the playback frequency;

[0027] The user stickiness rate is corrected based on the correction coefficient to obtain the actual user stickiness rate for each video type;

[0028] Determine the user profile of the video type corresponding to the maximum user stickiness rate, and use the user profile of the video type corresponding to the maximum user stickiness rate as the second user profile of the initial playback platform.

[0029] In one embodiment, determining the first user profile for each of the video types includes:

[0030] Obtain the user gender for each video type and the corresponding user consumption habits for each user gender;

[0031] Based on the user's gender and the user's consumption habits, determine the preferred product categories corresponding to each user's gender;

[0032] Obtain user age information corresponding to each of the preferred product categories, wherein the user age information includes the user's age group and the number of users in each age group;

[0033] Obtain the weight value of each user's age group, and determine the target age group for each preferred product category based on the number of users in each age group and the weight value.

[0034] The user's gender and target age group for each of the preferred product categories are determined as the first user profile for each of the video categories.

[0035] In one embodiment, determining the compatibility between the advertisement to be delivered and each of the initial playback platforms based on the advertisement type and the second user profile of each initial playback platform includes:

[0036] Based on the correspondence between the advertised type, the advertised type and the user profile, a third user profile corresponding to the advertised type is determined;

[0037] The third user profile is converted into a first feature vector, and the second user profile is converted into a second feature vector;

[0038] Calculate the similarity between the first feature vector and the second feature vector, and determine the similarity as the compatibility between the advertisement to be delivered and each of the initial playback platforms.

[0039] In one embodiment, determining a plurality of actual playback platforms based on the advertising quote includes:

[0040] Obtain the first user traffic of multiple platforms to be played within a preset time period, and determine the first playback platform from the multiple platforms to be played based on the first user traffic and the first traffic threshold of each platform to be played.

[0041] Obtain the second user traffic corresponding to the application scenario in each of the first playback platforms, and determine the second playback platform based on the second user traffic and the second traffic threshold;

[0042] Obtain revenue information for each of the second playback platforms;

[0043] Based on the revenue information and the advertising quote, determine the compatibility between the advertisement to be placed and each of the second playback platforms;

[0044] The actual playback platform is determined based on the compatibility of each of the second playback platforms and the preset number of required playback platforms.

[0045] Secondly, this application provides an advertising delivery optimization device based on artificial intelligence collaboration, which adopts the following technical solution:

[0046] An AI-based collaborative advertising optimization device includes:

[0047] The first acquisition module is used to, in response to receiving an advertising delivery task, acquire the advertising content of the advertisement to be delivered, and determine the advertising type of the advertisement to be delivered based on the advertising content;

[0048] The second acquisition module is used to acquire the advertising price of the advertisement to be placed, and determine several actual playback platforms based on the advertising price;

[0049] The target playback platform determination module is used to obtain the video type, playback duration and playback time of each actual playback platform, and match the advertisement to be delivered with each actual playback platform based on the advertisement type and the playback duration of each video type to obtain the target playback platform.

[0050] The delivery strategy determination module is used to determine the delivery strategy of the advertisement to be delivered based on the playback time of each video type.

[0051] The delivery module is used to deliver the advertisement to be delivered to the target playback platform based on the delivery strategy.

[0052] Thirdly, this application provides an electronic device that adopts the following technical solution:

[0053] At least one processor;

[0054] Memory;

[0055] At least one application, stored in the memory, when executed by the at least one processor, causes the at least one processor to perform: executing the AI-based collaborative ad delivery optimization method as described in any of the first aspects.

[0056] Fourthly, this application provides a computer-readable storage medium, which adopts the following technical solution:

[0057] A computer-readable storage medium having a computer program stored thereon, which, when executed in a computer, causes the computer to perform an AI-based collaborative advertising optimization method as described in any of the first aspects.

[0058] In summary, this application includes the following beneficial technical effects:

[0059] In response to receiving an advertising delivery task, the system determines the ad type based on the acquired ad content and the actual playback platform based on the acquired ad quote. It then obtains the video types, playback durations, and playback times for each actual playback platform and matches the ad type with the video type's playback duration to determine the target playback platform. A higher match between the ad type and the playback platform leads to a higher audience reach for the ad. Conversely, excessively long playback durations may reduce the audience reach. Therefore, determining the target playback platform based on the ad type and the playback duration of each video type effectively improves the accuracy of target playback platform determination. Finally, the system determines the delivery strategy based on the playback times of each video type and delivers the ad to the platform, effectively improving the ad delivery performance. Attached Figure Description

[0060] Figure 1 A flowchart illustrating an AI-based collaborative method for optimizing ad delivery is provided in this embodiment of the application.

[0061] Figure 2 A schematic diagram of an advertising delivery optimization device based on artificial intelligence collaboration, provided for an embodiment of this application;

[0062] Figure 3 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0063] The following is in conjunction with the appendix Figure 1 To be continued Figure 3 This application will be described in further detail.

[0064] This specific embodiment is merely an explanation of this application and is not intended to limit it. After reading this specification, those skilled in the art can make modifications to this embodiment without contributing any inventive step, but such modifications are protected by patent law as long as they are within the scope of this application.

[0065] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions in the embodiments of this application are described clearly and completely. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0066] Furthermore, the term "and / or" in this article is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. Additionally, the character " / " in this article, unless otherwise specified, generally indicates that the preceding and following related objects have an "or" relationship.

[0067] The embodiments of this application will now be described in further detail with reference to the accompanying drawings.

[0068] In today's rapidly developing digital economy, advertising has become one of the core means for businesses to enhance brand awareness, boost product sales, and expand market share. Whether it's online e-commerce platforms, social media, video websites, or offline smart screens and outdoor advertising spaces, advertising scenarios are becoming increasingly diverse, and audience reach paths are becoming more complex. For advertisers, the core demands of advertising always revolve around accurately targeting the right audience and maximizing return on investment—ensuring that advertising content reaches genuinely interested users with the means to pay, avoiding wasting resources on insignificant audiences, while maximizing advertising exposure within a limited budget.

[0069] Against this backdrop, embodiments of this application provide a technical solution that, in response to receiving an advertising delivery task, obtains the advertising content of the advertisement to be delivered, determines the advertising type based on the advertising content, and then matches the advertising type with the playback videos of each playback platform to obtain the target playback platform; obtains the advertising effectiveness information of the advertisement to be delivered, and determines the playback strategy of the advertisement to be delivered based on the advertising effectiveness information, the playback duration and playback period of each playback video, and delivers the advertisement to be delivered according to the playback strategy, effectively increasing the playback traffic of the advertisement to be delivered.

[0070] This application provides an AI-based collaborative advertising optimization method, executed by an electronic device. This electronic device can be a server or a terminal device. The server can be a standalone physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing cloud computing services. The terminal device can be a smartphone, tablet, laptop, desktop computer, etc., but is not limited to these. The terminal device and the server can be directly or indirectly connected via wired or wireless communication. This application does not impose any limitations on this. Figure 1 As shown, the method includes steps S1, S2, S3, S4, and S5, wherein:

[0071] Step S1: In response to receiving an ad delivery task, obtain the ad content of the ad to be delivered, and determine the ad type of the ad to be delivered based on the ad content.

[0072] Specifically, the advertising content includes advertising images and advertising copy; the advertising types include a first advertising type and a second advertising type. The first advertising type describes the application scenario of the advertisement to be placed. For example, if the application scenario of the advertisement to be placed is a kitchen, then the corresponding first advertising type is a kitchen waste and tableware advertisement; if the application scenario is a mountain climbing or hiking scenario, then the corresponding first advertising type can be an outdoor advertisement; the second advertising type describes the attributes of the advertisement to be placed. The second advertising type can be a product introduction advertisement, a promotional conversion advertisement, or a brand image advertisement, etc.

[0073] The electronic device has a pre-integrated monitoring program that monitors the triggering behavior of advertising delivery tasks. Once a delivery request is detected, an acquisition operation is performed. This includes directly extracting the data packet of the advertisement to be delivered, and then analyzing the extracted data packet to obtain the advertisement content.

[0074] The following example illustrates how determining the ad type based on ad content can be achieved. It is understood that ad images and text are the direct carriers of ad type, directly reflecting the differences between various ads. Determining the ad type based on ad content and copy can also reduce information loss and improve the accuracy of ad type determination; therefore, ad type should be determined based on ad content.

[0075] Step S2: Obtain the advertising quote for the ad to be placed, and determine several actual playback platforms based on the advertising quote.

[0076] Specifically, advertising quotes can be obtained from an advertising database. The specific process for determining the playback platform can be referred to in the following embodiments. In the embodiments of this application, the playback platform can be a text-based playback platform, such as Toutiao, WeChat official accounts, etc.; it can also be a short video-based playback platform, such as Xigua Video, Douyin Video, etc.; or it can be a long video-based playback platform, such as iQiyi, local TV stations, etc.

[0077] Step S3: Obtain the video types, playback durations, and playback periods of each actual playback platform, and match the advertisements to be delivered with each actual playback platform based on the advertisement type and the playback duration of each video type to obtain the target playback platform.

[0078] Specifically, the video can be a long video, such as a TV series or a movie, or a short video. The video type can be comedy, family, parenting, or science, etc., and this application embodiment does not limit this. The video type, playback duration, and playback time period can be obtained from the playback platform's information statistics database. The specific process of matching the advertisement to be delivered based on the advertisement type and each playing video to obtain the target playback platform can be referred to in the following embodiment. It is understood that matching the advertisement type and the playback duration of each video type allows the advertisement to be delivered to be embedded in a highly compatible viewing scenario for the user, avoiding a disconnect between the advertisement and the scene, while also avoiding resource waste and maximizing the complete playback rate of the advertisement.

[0079] Step S4: Determine the placement strategy for the ads to be placed based on the playback time of each video type.

[0080] Specifically, the playback period of the advertisement to be placed is determined according to the playback period and set playback duration of each video type, and the advertisement to be placed is played within the playback period of the advertisement to be placed.

[0081] Step S5: Deploy the advertisement to be delivered on the target playback platform based on the delivery strategy.

[0082] Specifically, the ads to be delivered are played at a set frequency on the target playback platform according to the set playback duration and time period, so as to optimize ad delivery.

[0083] Based on the above embodiments, in response to receiving an advertising delivery task, the system determines the advertising type based on the obtained advertising content and the actual playback platform based on the obtained advertising quote, so that the advertising quote can be played on the actual playback platform. It then obtains the video type, playback duration, and playback time period for each actual playback platform, and matches the advertising type and the playback duration of the video type to determine the target playback platform. The higher the compatibility between the advertising type and the playback platform, the higher the audience reach of the advertisement to be delivered. However, if the playback duration of the advertisement to be delivered is too long, it may lead to a decrease in the audience reach of the advertisement to be delivered. Therefore, it is necessary to determine the target playback platform based on the advertising type and the playback duration of each video type to effectively improve the accuracy of the target playback platform determination. Finally, the system determines the delivery strategy based on the playback time period of each video type and delivers the advertisement to be delivered, thereby effectively improving the delivery effect of the advertisement to be delivered.

[0084] One possible implementation of this application embodiment is that the advertising content includes advertising images and advertising copy, and the type of advertising to be delivered is determined based on the advertising content, including:

[0085] The advertisement image is input into a pre-trained advertisement recognition model to obtain the first advertisement type of the advertisement to be delivered. The first advertisement type is used to describe the application scenario of the advertisement to be delivered.

[0086] Obtain several keywords corresponding to multiple preset second ad types. The preset second ad types are used to describe the attributes of the ad to be delivered.

[0087] Match each keyword with the advertising copy to obtain the target keywords for the advertisement to be placed;

[0088] Determine the keyword frequency of each target keyword, and determine the target secondary ad type corresponding to the ad to be placed based on the keyword frequency of each target keyword.

[0089] Specifically, the pre-trained advertising recognition model is trained based on multiple historical data and a neural network model. This application embodiment does not limit the specific training process of the advertising recognition model; users can set it themselves. Furthermore, it is understood that using a pre-trained advertising recognition model to identify advertising images can effectively improve the accuracy and efficiency of advertising image recognition. Multiple keywords corresponding to preset second advertising types can be obtained from a preset database. Each preset second advertising type can correspond to multiple keywords; for example, when the preset second advertising type is a product introduction advertisement, the corresponding keywords can be "clearing dead corners, solving family problems, 1800Pa strong suction," etc.; when the preset second advertising type is a promotional advertisement, the corresponding keywords can be "limited-time discount, low inventory, today only," etc.; when the preset second advertising type is an image-building advertisement, the corresponding keywords can be "rural education, practicing brand concepts," etc. The keyword matching algorithm matches each keyword with the advertising copy to obtain the target keywords in the advertising copy of the advertisement to be placed. This application embodiment does not limit the specific process of determining the frequency of keyword occurrences; users can set it themselves. The specific process of determining the preset second advertising type based on the frequency of occurrence of each target keyword includes: selecting the target keyword corresponding to the highest keyword frequency; comparing the highest keyword frequency with a keyword frequency threshold; if the highest keyword frequency is greater than the preset keyword frequency threshold, then the preset second advertising type corresponding to the target keyword with the highest keyword frequency is determined as the advertising type of the advertisement to be placed; if the highest keyword frequency is not greater than the preset keyword frequency threshold, then a semantic recognition algorithm is used to identify the semantics of the advertising copy, and then the semantics of the advertising copy are matched with the advertisements corresponding to each second advertising type to determine the target second advertising type corresponding to the advertisement to be placed. It is understood that first determining the second advertising type of the advertisement to be placed based on the keyword frequency can reduce the determination time of the second advertising type and improve recognition efficiency.

[0090] Based on the above embodiments, the advertising image is input into a pre-trained advertising recognition model to obtain the first advertising type of the advertisement to be placed, several keywords corresponding to multiple preset second advertising types are obtained, and the target keywords are obtained by matching the keywords with the advertising copy. Then, the target second advertising type of the advertisement to be placed is determined by the frequency of occurrence of the target keywords, so as to effectively improve the accuracy of the determination of the second advertising type.

[0091] One possible implementation of this application embodiment involves matching the advertisement to be delivered with various actual playback platforms to obtain the target playback platform based on the advertisement type and the playback duration of each video type, including:

[0092] Based on the playback duration of each video type, determine the average playback duration of each actual playback platform;

[0093] Obtain the playback duration of the ad to be delivered, and compare the playback duration of the ad to be delivered with the average playback duration to determine the initial playback platform. The average playback duration of the initial playback platform is greater than the playback duration of the ad to be delivered.

[0094] Determine the first user profile for each video category;

[0095] The playback frequency of each video type is obtained, and the second user profile of each initial playback platform is determined based on the playback frequency and the first user profile corresponding to each video type.

[0096] The suitability of the advertisements to be placed with each primary actual platform is determined based on the advertisement type and the second user profile of each initial playback platform.

[0097] Determine the target playback platform based on compatibility;

[0098] Specifically, based on the first user profile corresponding to each playback frequency and video type, the second user profile of each initial playback platform is determined, including:

[0099] For each initial playback platform, obtain the user stickiness rate corresponding to each video type;

[0100] Based on the correspondence between the initial playback platform playback frequency, playback frequency, and correction coefficient, determine the correction coefficient corresponding to the initial playback platform playback frequency;

[0101] The user stickiness rate of the initial playback platform is corrected based on the initial playback platform correction coefficient to obtain the actual user stickiness rate of each video type.

[0102] Determine the user profile of the video type corresponding to the maximum user stickiness rate, and use the user profile of the video type corresponding to the maximum user stickiness rate as the second user profile of the initial playback platform.

[0103] Specifically, the average playback duration can be determined based on the number of video types and the total playback duration. This application does not limit the specific process for determining the average playback duration. The playback duration of the advertisements to be delivered can be obtained from the advertisement information database. It is understood that when the playback duration of the advertisements to be delivered is too long, i.e., higher than the average playback duration of the actual playback platform, the completion rate may decrease. Under the influence of the platform's streaming mechanism, when the completion rate decreases, the streaming of the advertisements to be delivered decreases, which may affect the audience reach of the advertisements. Therefore, it is necessary to consider both the playback duration of the advertisements to be delivered and the average playback duration of the actual playback platform, and determine the actual playback platform with an average playback duration greater than the playback duration of the advertisements to be delivered as the initial playback platform for the advertisements to be delivered. The specific process for determining the first user profile corresponding to each video type can be referred to in the following embodiments. It is understood that the essence of determining the first user profile through video type is to accurately bind content attributes and user needs, making the video types more targeted. Furthermore, for each initial playback platform, the playback frequency corresponding to each video type can be obtained from the playback platform information database. The specific process of determining the second user profile for each playback platform based on playback frequency and the first user profile corresponding to each video type includes: User stickiness rate can be obtained from the video data statistics database. User stickiness rate describes the proportion of users who watch this video type daily within a preset duration, out of all users watching this video type. The correspondence between playback frequency and correction coefficient is preset by technical personnel. The correction coefficient is used to correct the user stickiness rate for this video type. The correction can be performed using the following formula: Actual User Stickiness Rate = User Stickiness Rate * Correction Coefficient. By comparing the actual user stickiness rates of each video type, the user profile corresponding to the highest user stickiness rate can be determined as the second user profile for the playback platform.

[0104] It is understandable that each initial playback platform includes multiple video types, meaning each initial playback platform includes multiple first user profiles. To more quickly match the actual playback platform with the advertisements to be delivered, it is necessary to determine targeted second user profiles. The specific process of determining the fit based on the advertisement type and the second user profiles of each initial playback platform can be referred to in the following embodiment, where fit describes the degree of matching between the advertisement type and the second user profile. The higher the fit between the advertisement type and the second user profile, the more similar the advertisement type is to the second user profile. A higher similarity between the advertisement type and the second user profile indicates that the advertisement to be delivered better matches the consumption preferences and habits of the users corresponding to the second user profile. By delivering the advertisement on this initial playback platform, compared to other initial playback platforms, the initial playback platform with a higher fit with the advertisement type has a better advertising effect. Therefore, it is necessary to determine the fit between the advertisement type and each initial playback platform to effectively improve the effectiveness of the advertisement delivery. Furthermore, the specific process of determining the target playback platform based on the adaptation degree includes: selecting the initial playback platform corresponding to the maximum adaptation degree, and determining the initial playback platform corresponding to the maximum adaptation degree as the target playback platform.

[0105] Based on the above embodiments, playback duration affects the audience acquisition rate of the advertisement to be delivered. Therefore, it is necessary to determine the average playback duration of the actual playback platform and compare the playback duration of the advertisement to be delivered with the average playback duration to determine the initial playback platform with a playback duration longer than that of the advertisement to be delivered, so as to avoid a decrease in audience acquisition due to excessively long playback duration. Then, a first user profile is determined based on the video type, and a second user profile of the initial playback platform is determined based on the playback frequency of the video type and the first user profile. The suitability between the advertisement type and the initial playback platform is determined based on the advertisement type and the second user profile, and the target playback platform is determined based on the suitability, which effectively improves the accuracy of the target playback platform determination.

[0106] One possible implementation of this application embodiment involves determining a first user profile for each video type, including:

[0107] Obtain the user gender for each video category and the corresponding user consumption habits for each gender;

[0108] Based on user gender and user consumption habits, determine the preferred product categories corresponding to each user's gender;

[0109] Obtain user age information corresponding to each preferred product category. User age information includes the user's age group and the number of users in each age group.

[0110] Obtain the weight value of each user's age group, and determine the target age group for each preferred product category based on the number of users in each age group and each weight value;

[0111] The gender and target age group of users for each preferred product category are determined as the first user profile for each video category.

[0112] Specifically, user consumption habits represent the product categories with the highest purchase quantity or the highest purchase expenditure. User gender and corresponding consumption habits can be obtained from the video information database of each played video. The specific process of determining the preferred product categories for each user gender based on their gender and consumption habits includes: for the same user gender, obtaining the frequency of occurrence of each preferred product category, arranging the preferred product categories in descending order of frequency, and selecting a preset number of preferred product categories as the preferred product categories corresponding to that user gender. User age information can be obtained from the user information database corresponding to the product categories, where each age group can be divided into ten-year segments or twenty-year segments. In this embodiment, the age groups corresponding to different preferred product categories are different; for example, the age group corresponding to maternal and infant products may be 35-45 years old, and the age group corresponding to health products may be 55-70 years old. The weight values ​​for each user's age group are preset by relevant technical personnel. The specific process of determining the target age group based on the number of users and each weight value includes: determining the target correction coefficient corresponding to the number of users based on the correspondence between the number of users, the number of users, and the correction coefficient; using the target correction coefficient to correct the weight values ​​to obtain the corrected weight values; selecting the age group of users corresponding to the maximum weight value, and determining the age group of users corresponding to the maximum weight value as the target age group. Further, the first user profile corresponding to the video type includes preferred product categories, user gender, and the target age group corresponding to each user gender.

[0113] Based on the above embodiments, user gender and corresponding user consumption habits are obtained, and preferred product categories are accurately located based on user consumption habits; user age information corresponding to preferred product categories and weight values ​​of user age groups are obtained, and target age groups for preferred product categories are determined based on user numbers and weight values, so as to achieve accurate determination of target age groups. Then, the user gender and target age groups for preferred product categories are determined as the first user profile, which effectively improves the accuracy of the first profile determination.

[0114] One possible implementation of this application embodiment involves determining the compatibility between the advertisement to be delivered and each initial playback platform based on the advertisement type and the second user profile of each initial playback platform, including:

[0115] Based on the correspondence between ad type, ad type and user profile, determine the third user profile corresponding to the ad type;

[0116] The third user profile is transformed into the first feature vector, and the second user profile is transformed into the second feature vector;

[0117] Calculate the similarity between the first feature vector and the second feature vector, and determine the similarity as the compatibility between the advertisement to be delivered and each initial playback platform.

[0118] Specifically, the advertising type and the user profile are pre-defined by technical personnel. The user profile represents the user profile of the users expected to be targeted by the advertising. The specific process of converting the third user profile into the first feature vector includes: obtaining the profile features in the third user profile, and quantifying the profile features according to the profile features, the profile features, and the corresponding quantization rules. For example, when the profile features of the third user profile include monthly consumption amount and user age group, which are 500 and 26 years old respectively, the quantization rule corresponding to monthly consumption amount is 1 when monthly consumption amount is 100-600 yuan, and 0.8 when monthly consumption amount is 600-1500 yuan. The quantization rule corresponding to user age group is 1 when user age group is 22-28 years old, 0.8 when user age group is 29-35 years old, and 0.5 when user age group is 36-45 years old. Further, the first feature vector can be determined as [1,1]. The process of determining the second feature vector is the same as the process of determining the first feature vector, and will not be repeated in the embodiments of this application. Furthermore, the similarity between the first and second feature vectors can be calculated using the cosine similarity formula or the Euclidean distance formula, and the similarity can be determined as the fitness.

[0119] Based on the above embodiments, a third user profile is determined according to the type of advertisement. The third user profile and the second user profile are then converted into a first feature vector and a second feature vector, respectively, for easy calculation. The similarity is then calculated based on the feature vectors, and the similarity is determined as the fit, which effectively improves the accuracy of the fit determination.

[0120] One possible implementation of this application embodiment involves determining several actual playback platforms based on advertising quotes, including:

[0121] Get the first user traffic of multiple platforms to be played within a preset time period, and determine the first playback platform from the multiple platforms to be played based on the first user traffic and the first traffic threshold of each platform to be played.

[0122] Obtain the second user traffic corresponding to the application scenario in each first playback platform, and determine the second playback platform based on the second user traffic and the second traffic threshold;

[0123] Obtain revenue information from various secondary playback platforms;

[0124] Based on revenue information and advertising rates, determine the compatibility between the advertisements to be placed and each secondary playback platform;

[0125] The actual playback platforms are determined based on the compatibility of each secondary playback platform and the number of preset required playback platforms.

[0126] Specifically, the first user traffic represents the number of users entering the platform to be played; the preset duration is pre-set by technical personnel. The first user traffic is compared with a first traffic threshold, and platforms with first user traffic exceeding the first traffic threshold are identified as the first playback platforms. From the first playback platforms, the application scenario corresponding to the advertisement to be placed is determined, and the second user traffic, which describes the number of users who viewed videos of that application scenario, is obtained from the user information database. The second user traffic is then compared with the second user traffic threshold, and platforms with second user traffic exceeding the second user traffic threshold are identified as the second playback platforms. Here, the preset duration, the first traffic threshold, and the second traffic threshold are all pre-set by technical personnel; the first traffic threshold can be the same or different for different platforms to be played. Revenue information describes the revenue of the second playback platform within the preset duration and can be obtained from the platform information database. The specific process of determining the fit based on the revenue information and the advertising price includes: calculating the return on investment (ROI) for each second playback platform according to the ROI calculation formula, and determining the ROI as the fit based on the correspondence between ROI and fit. The number of preset playback platforms is set by the technicians, and a second playback platform can be selected from the preset number of playback platforms.

[0127] Based on the above embodiments, the platforms to be played are initially screened based on the first user traffic of the platforms to be played, and then screened again based on the second user traffic of the platforms to be advertised, so as to obtain the second playback platforms with high user traffic; then the revenue information of the second playback platforms is obtained to determine the playback platforms with higher revenue information from the high user traffic platforms; the suitability is determined based on the revenue information and the advertising price, so as to accurately screen the second playback platforms with high user traffic and high revenue; then the actual playback platforms of the preset number of playback platforms are selected. The playback platforms are screened from two dimensions: user traffic and revenue information, which effectively improves the accuracy of playback platform screening.

[0128] The above embodiments describe an AI-based collaborative advertising optimization method from the perspective of process flow. The following embodiments describe an AI-based collaborative advertising optimization device from the perspective of virtual modules or virtual units. For details, please refer to the following embodiments.

[0129] This application provides an advertising delivery optimization device based on artificial intelligence collaboration, such as... Figure 2As shown, the AI-based collaborative advertising optimization device may specifically include:

[0130] The first acquisition module 201 is used to, in response to receiving an advertising delivery task, acquire the advertising content of the advertisement to be delivered, and determine the advertising type of the advertisement to be delivered based on the advertising content;

[0131] The second acquisition module 202 is used to acquire the advertising price of the advertisement to be placed, and determine several actual playback platforms based on the advertising price;

[0132] The target playback platform determination module 203 is used to obtain the video types, playback durations and playback periods of each actual playback platform, and to match the advertisement to be delivered with each actual playback platform based on the advertisement type and the playback duration of each video type to obtain the target playback platform.

[0133] The delivery strategy determination module 204 is used to determine the delivery strategy of the advertisement to be delivered based on the playback time of each video type.

[0134] The delivery module 205 is used to deliver the advertisement to be delivered on the target playback platform based on the delivery strategy.

[0135] Based on the above embodiments, in response to receiving an advertising delivery task, the system determines the advertising type based on the obtained advertising content and the actual playback platform based on the obtained advertising quote, so that the advertising quote can be played on the actual playback platform. It then obtains the video type, playback duration, and playback time period for each actual playback platform, and matches the advertising type and the playback duration of the video type to determine the target playback platform. The higher the compatibility between the advertising type and the playback platform, the higher the audience reach of the advertisement to be delivered. However, if the playback duration of the advertisement to be delivered is too long, it may lead to a decrease in the audience reach of the advertisement to be delivered. Therefore, it is necessary to determine the target playback platform based on the advertising type and the playback duration of each video type to effectively improve the accuracy of the target playback platform determination. Finally, the system determines the delivery strategy based on the playback time period of each video type and delivers the advertisement to be delivered, thereby effectively improving the delivery effect of the advertisement to be delivered.

[0136] In one possible implementation of this application embodiment, the advertising content includes advertising images and advertising text. When the first acquisition module 201 determines the type of advertisement to be delivered based on the advertising content, it is used to:

[0137] The advertisement image is input into a pre-trained advertisement recognition model to obtain the first advertisement type of the advertisement to be delivered. The first advertisement type is used to describe the application scenario of the advertisement to be delivered.

[0138] Obtain several keywords corresponding to multiple preset second ad types. The preset second ad types are used to describe the attributes of the ad to be delivered.

[0139] Match each keyword with the advertising copy to obtain the target keywords for the advertisement to be placed;

[0140] Determine the keyword frequency of each target keyword, and determine the target secondary ad type corresponding to the ad to be placed based on the keyword frequency of each target keyword.

[0141] In one possible implementation of this application embodiment, when the first acquisition module 201 performs the matching of the advertisement to be delivered with each actual playback platform based on the advertisement type and the playback duration of each video type to obtain the target playback platform, it is used to:

[0142] Based on the playback duration of each video type, determine the average playback duration of each actual playback platform;

[0143] Obtain the playback duration of the ad to be delivered, and compare the playback duration of the ad to be delivered with the average playback duration to determine the initial playback platform. The average playback duration of the initial playback platform is greater than the playback duration of the ad to be delivered.

[0144] Determine the first user profile for each video category;

[0145] The playback frequency of each video type is obtained, and the second user profile of each initial playback platform is determined based on the playback frequency and the first user profile corresponding to each video type.

[0146] The suitability of the advertisement to be delivered to each initial playback platform is determined based on the advertisement type and the second user profile of each initial playback platform.

[0147] Determine the target playback platform based on compatibility.

[0148] In one possible implementation of this application embodiment, when the first acquisition module 201 executes the process of determining the second user profile of each initial playback platform based on the first user profile corresponding to each playback frequency and each video type, it is used to:

[0149] For each initial playback platform, obtain the user stickiness rate corresponding to each video type;

[0150] Based on the correspondence between playback frequency, playback frequency and correction coefficient, determine the correction coefficient corresponding to the playback frequency;

[0151] The user stickiness rate is corrected based on the correction factor to obtain the actual user stickiness rate for each video type;

[0152] Determine the user profile for the video type corresponding to the maximum user stickiness rate, and use the user profile for the video type corresponding to the maximum user stickiness rate as the second user profile for the initial playback platform.

[0153] In one possible implementation of this application embodiment, when the first acquisition module 201 performs the process of determining the first user profile for each video type, it is used to:

[0154] Obtain the user gender for each video category and the corresponding user consumption habits for each gender;

[0155] Based on user gender and user consumption habits, determine the preferred product categories corresponding to each user's gender;

[0156] Obtain user age information corresponding to each preferred product category. User age information includes the user's age group and the number of users in each age group.

[0157] Obtain the weight value of each user's age group, and determine the target age group for each preferred product category based on the number of users in each age group and each weight value;

[0158] The gender and target age group of users for each preferred product category are determined as the first user profile for each video category.

[0159] In one possible implementation of this application embodiment, when the first acquisition module 201 determines the compatibility between the advertisement to be delivered and each initial playback platform based on the advertisement type and the second user profile of each initial playback platform, it is used to:

[0160] Based on the correspondence between ad type, ad type and user profile, determine the third user profile corresponding to the ad type;

[0161] The third user profile is transformed into the first feature vector, and the second user profile is transformed into the second feature vector;

[0162] Calculate the similarity between the first feature vector and the second feature vector, and determine the similarity as the compatibility between the advertisement to be delivered and each initial playback platform.

[0163] In one possible implementation of this application embodiment, when the second acquisition module 202 is executing the function of determining several actual playback platforms based on advertising quotes, it is used for:

[0164] Get the first user traffic of multiple platforms to be played within a preset time period, and determine the first playback platform from the multiple platforms to be played based on the first user traffic and the first traffic threshold of each platform to be played.

[0165] Obtain the second user traffic corresponding to the application scenario in each first playback platform, and determine the second playback platform based on the second user traffic and the second traffic threshold;

[0166] Obtain revenue information from various secondary playback platforms;

[0167] Based on revenue information and advertising rates, determine the compatibility between the advertisements to be placed and each secondary playback platform;

[0168] The actual playback platforms are determined based on the compatibility of each secondary playback platform and the number of preset required playback platforms.

[0169] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process of the above-described AI-based collaborative advertising optimization device can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0170] This application provides an electronic device, such as... Figure 3 As shown, Figure 3 The illustrated electronic device includes a processor 301 and a memory 303. The processor 301 and the memory 303 are connected, for example, via a bus 302. Optionally, the electronic device may also include a transceiver 304. It should be noted that in practical applications, the transceiver 304 is not limited to one type, and the structure of this electronic device does not constitute a limitation on the embodiments of this application.

[0171] Processor 301 may be a CPU (Central Processing Unit), a general-purpose processor, a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. It can implement or execute the various exemplary logic blocks, modules, and circuits described in conjunction with the disclosure of this application. Processor 301 may also be a combination that implements computational functions, such as including one or more microprocessor combinations, a combination of a DSP and a microprocessor, etc.

[0172] Bus 302 may include a pathway for transmitting information between the aforementioned components. Bus 302 may be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus, etc. Bus 302 can be divided into address bus, data bus, control bus, etc. For ease of representation, Figure 3 The symbol is represented by a single thick line, but this does not mean that there is only one bus or one type of bus.

[0173] The memory 303 may be a ROM (Read Only Memory) or other type of static storage device capable of storing static information and instructions, RAM (Random Access Memory) or other type of dynamic storage device capable of storing information and instructions, or an EEPROM (Electrically Erasable Programmable Read Only Memory), CD-ROM (Compact Disc Read Only Memory) or other optical disc storage, optical disc storage (including compressed optical discs, laser discs, optical discs, digital universal optical discs, Blu-ray discs, etc.), magnetic disk storage media or other magnetic storage devices, or any other medium capable of carrying or storing desired program code in the form of instructions or data structures and accessible by a computer, but not limited thereto.

[0174] The memory 303 is used to store application code that executes the solution of this application, and its execution is controlled by the processor 301. The processor 301 is used to execute the application code stored in the memory 303 to implement the content shown in the foregoing method embodiments.

[0175] Electronic devices include, but are not limited to: mobile terminals such as mobile phones, laptops, digital radio receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), and in-vehicle terminals (such as in-vehicle navigation terminals), as well as fixed terminals such as digital TVs and desktop computers. Servers can also be included. Figure 3 The electronic device shown is merely an example and should not impose any limitation on the functionality and scope of use of the embodiments of this application.

[0176] This application provides a computer-readable storage medium storing a computer program. When the program is run on a computer, it enables the computer to execute the corresponding content in the aforementioned method embodiments. Compared to related technologies, in response to receiving an advertising delivery task, the program determines the advertising type based on the obtained advertising content and the actual playback platform based on the obtained advertising price. It then obtains the video type, playback duration, and playback time period for each actual playback platform and matches the advertising type and video type playback duration to determine the target playback platform. The higher the compatibility between the advertising type and the playback platform, the higher the audience reach of the advertisement to be delivered. However, if the playback duration of the advertisement to be delivered is too long, it may lead to a decrease in the audience reach. Therefore, it is necessary to determine the target playback platform based on the advertising type and the playback duration of each video type to effectively improve the accuracy of target playback platform determination. Finally, the program determines the delivery strategy based on the playback time period of each video type and delivers the advertisement to be delivered, thereby effectively improving the delivery effect of the advertisement to be delivered.

[0177] It should be understood that although the steps in the flowcharts of the accompanying figures are shown sequentially as indicated by the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the accompanying figures may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times, and their execution order is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the sub-steps or stages of other steps.

[0178] The above description is only a partial embodiment of this application. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of this application, and these improvements and modifications should also be considered within the scope of protection of this application.

Claims

1. An advertising delivery optimization method based on artificial intelligence collaboration, characterized in that, include: In response to receiving an advertising delivery task, the system obtains the advertising content of the advertisement to be delivered and determines the advertising type of the advertisement to be delivered based on the advertising content. Obtain the advertising quote for the advertisement to be placed, and determine several actual playback platforms based on the advertising quote; The video types, playback durations, and playback periods of each actual playback platform are obtained, and based on the ad type and the playback duration of each video type, the ad to be delivered is matched with each actual playback platform to obtain the target playback platform. The placement strategy for the advertisements to be placed is determined based on the playback time of each video type. The advertisement to be advertised will be advertised on the target playback platform based on the aforementioned advertising strategy. The advertising content includes advertising images and advertising copy. Determining the advertising type of the advertisement to be delivered based on the advertising content includes: The advertisement image is input into a pre-trained advertisement recognition model to obtain the first advertisement type of the advertisement to be delivered. The first advertisement type is used to describe the application scenario of the advertisement to be delivered. Obtain several keywords corresponding to multiple preset second ad types, wherein the preset second ad types are used to describe the attributes of the ad to be placed; The keywords are matched with the advertising copy to obtain the target keywords for the advertisement to be placed. Determine the keyword frequency of each target keyword, and determine the target second ad type corresponding to the ad to be delivered based on the keyword frequency of each target keyword; The step of matching the advertisement to be delivered with each of the actual playback platforms based on the advertisement type and the playback duration of each of the video types to obtain the target playback platform includes: Based on the playback duration of each video type, determine the average playback duration of each actual playback platform; The playback duration of the advertisement to be delivered is obtained, and the playback duration of the advertisement to be delivered is compared with the average playback duration to determine the initial playback platform, wherein the average playback duration of the initial playback platform is greater than the playback duration of the advertisement to be delivered. Determine the first user profile for each of the aforementioned video types; The playback frequency of each video type is obtained, and the second user profile of each initial playback platform is determined based on the playback frequency and the first user profile corresponding to each video type. The compatibility between the advertisement to be delivered and each of the initial playback platforms is determined based on the advertisement type and the second user profile of each initial playback platform. The target playback platform is determined based on the compatibility score. The step of determining the compatibility between the advertisement to be delivered and each of the initial playback platforms based on the advertisement type and the second user profile of each initial playback platform includes: Based on the correspondence between the advertised type, the advertised type and the user profile, a third user profile corresponding to the advertised type is determined; The third user profile is converted into a first feature vector, and the second user profile is converted into a second feature vector; Calculate the similarity between the first feature vector and the second feature vector, and determine the similarity as the compatibility between the advertisement to be delivered and each of the initial playback platforms.

2. The advertising delivery optimization method based on artificial intelligence collaboration as described in claim 1, characterized in that, The step of determining the second user profile of each initial playback platform based on the first user profile corresponding to each playback frequency and each video type includes: For each initial playback platform, obtain the user stickiness rate corresponding to each video type; Based on the correspondence between the playback frequency, the playback frequency and the correction coefficient, determine the correction coefficient corresponding to the playback frequency; The user stickiness rate is corrected based on the correction coefficient to obtain the actual user stickiness rate for each video type; Determine the user profile of the video type corresponding to the maximum user stickiness rate, and use the user profile of the video type corresponding to the maximum user stickiness rate as the second user profile of the initial playback platform.

3. The advertising delivery optimization method based on artificial intelligence collaboration as described in claim 1, characterized in that, The determination of the first user profile for each of the aforementioned video types includes: Obtain the user gender for each video type and the corresponding user consumption habits for each user gender; Based on the user's gender and the user's consumption habits, determine the preferred product categories corresponding to each user's gender; Obtain user age information corresponding to each of the preferred product categories, wherein the user age information includes the user's age group and the number of users in each age group; Obtain the weight value of each user's age group, and determine the target age group for each preferred product category based on the number of users in each age group and the weight value. The user's gender and target age group for each of the preferred product categories are determined as the first user profile for each of the video categories.

4. The advertising delivery optimization method based on artificial intelligence collaboration as described in claim 1, characterized in that, The determination of several actual playback platforms based on the advertising quote includes: Obtain the first user traffic of multiple platforms to be played within a preset time period, and determine the first playback platform from the multiple platforms to be played based on the first user traffic and the first traffic threshold of each platform to be played. Obtain the second user traffic corresponding to the application scenario in each of the first playback platforms, and determine the second playback platform based on the second user traffic and the second traffic threshold; Obtain revenue information for each of the second playback platforms; Based on the revenue information and the advertising quote, determine the compatibility between the advertisement to be placed and each of the second playback platforms; The actual playback platform is determined based on the compatibility of each of the second playback platforms and the preset number of required playback platforms.

5. An advertising delivery optimization device based on artificial intelligence collaboration, characterized in that, The application of the AI-based collaborative advertising delivery optimization method as described in any one of claims 1 to 4 includes: The first acquisition module is used to, in response to receiving an advertising delivery task, acquire the advertising content of the advertisement to be delivered, and determine the advertising type of the advertisement to be delivered based on the advertising content; The second acquisition module is used to acquire the advertising price of the advertisement to be placed, and determine several actual playback platforms based on the advertising price; The target playback platform determination module is used to obtain the video type, playback duration and playback time of each actual playback platform, and match the advertisement to be delivered with each actual playback platform based on the advertisement type and the playback duration of each video type to obtain the target playback platform. The delivery strategy determination module is used to determine the delivery strategy of the advertisement to be delivered based on the playback time of each video type. The delivery module is used to deliver the advertisement to be delivered to the target playback platform based on the delivery strategy.

6. An electronic device, characterized in that, include: At least one processor; Memory; At least one application is stored in the memory, and when the at least one application is executed by the at least one processor, the at least one processor performs the AI-based collaborative advertising optimization method according to any one of claims 1 to 4.

7. A computer-readable storage medium, characterized in that, It stores a computer program, which, when executed in a computer, causes the computer to perform the advertising placement optimization method based on artificial intelligence collaboration as described in any one of claims 1 to 4.

Images