A multimedia information system for advertising design

Abstract

The present invention discloses an advertisement design multimedia information system, which relates to the technical field of advertisement design. The present invention comprises: an advertisement design unit, a publishing monitoring unit and a database. The advertisement design unit comprises a template analysis module and an advertisement generation module. The publishing monitoring unit comprises an advertisement publishing module, a publishing effect monitoring module and an advertisement promotion analysis module. By analyzing the duration, background music and video special effects of video advertisements corresponding to target users, a video advertisement is generated, an appropriate delivery time is selected for delivery, the effect of the advertisement delivery is monitored, and corresponding promotion is carried out according to the delivery effect. The intelligent production and delivery of video advertisements is realized, the attractiveness of video advertisements and the public's acceptance are improved, the audience's viewing experience is enhanced, the promotion effect, exposure rate and conversion rate of video advertisements are guaranteed, and the promotion cost of advertisements is also reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of advertisement design, and in particular to an advertisement design multimedia information system. Background Art

[0002] Short videos, with their concise and powerful nature, meet the needs of modern people for the fragmented time in their fast-paced lives. Short video platforms have a vast user base encompassing all ages and interests, and their user base is highly active. Therefore, precisely designing and placing video ads based on product characteristics and the characteristics of the target audience on the short video platform can increase brand awareness and exposure.

[0003] Existing technologies such as the application with publication number CN115760020A disclose an Internet-based video integrated media content analysis and management system, which includes a candidate material video extraction module, a candidate material video content analysis module, a candidate material video screening module, a city scenic spot operation benefit acquisition module, a target city promotion video splicing module, a candidate background music combination module, a candidate background music screening module and a city information storage library; the present invention improves the credibility and authenticity of the city scenic spot promotion video by screening the specified material video from the candidate material videos corresponding to each city scenic spot, and splicing them into a promotional video of the target city, so that the city scenic spot promotion can achieve the expected effect, and combines each candidate background music with the promotional video of the target city to screen out the best promotional video of the target city, which effectively meets the cultural needs of the general public.

[0004] The above technical solution has at least the following deficiencies: 1. Short video platforms feature short duration videos. Therefore, it is necessary to select appropriate video advertisement duration based on the viewing rate of videos of various durations by the target group of the advertisement. The above solution lacks analysis of video duration, and thus cannot guarantee the suitability of video advertisement duration, thus affecting the effectiveness of advertising.

[0005] 2. Background music and video effects in short videos can not only improve the quality, expressiveness and appeal of the video, but also expand the video's dissemination channels and audience groups, and enhance the audience's viewing experience. Therefore, choosing suitable background music and video effects for video advertisements can increase the advertising dissemination effect. However, the selection of background music in the above scheme is based on the beat and style of the music. At the same time, there is a lack of selection of video effects, and no analysis of the novelty of background music and video effects and the acceptance of the target group. As a result, it is impossible to improve the appeal of subsequent videos and the public's acceptance, and it is impossible to reflect the creativity of the advertisement, and it is impossible to improve the audience's viewing experience, which reduces the publicity effect of the advertisement.

[0006] 3. Different groups in short videos have different levels of activity at different times. Therefore, when conducting advertising promotion, it is necessary to select appropriate promotion time periods based on the online status of the account's own fan group and the online status of the user group on the short video platform. This can increase advertising traffic. However, the above solution lacks analysis of the video's promotion time period, and thus cannot guarantee the promotion effect, exposure rate, and conversion rate of video advertising, nor can it reduce the promotion cost of advertising. Summary of the Invention

[0007] In view of the above-mentioned technical deficiencies, the purpose of the present invention is to provide an advertisement design multimedia information system.

[0008] In order to solve the above technical problems, the present invention adopts the following technical solution: The present invention provides an advertisement design multimedia information system, including: an advertisement design unit and a release monitoring unit.

[0009] The advertisement design unit includes a template analysis module and an advertisement generation module;

[0010] The template analysis module is used to obtain the element data corresponding to the product after the target user uploads the product material data, and analyze the template of the video advertisement corresponding to the product. The template includes the recommended duration, recommended background music and recommended video special effects template.

[0011] The advertisement generation module is used to import product material data into the template of the product's corresponding video advertisement to generate the product's corresponding video advertisement.

[0012] The publishing monitoring unit includes an advertisement publishing module, a publishing effect monitoring module and an advertisement promotion analysis module.

[0013] The advertising publishing module is used to obtain the historical video publishing data corresponding to the target user account, and obtain the target group data corresponding to the product, analyze the recommended sending time of the product's video advertisement, and provide prompts.

[0014] The release effect monitoring module is used to arrange each monitoring moment according to the preset time interval after the product video advertisement is sent, collect the promotional data corresponding to each monitoring moment, and analyze the product advertising release effect.

[0015] The advertising promotion analysis module is used to utilize the advertising release effect of the product, analyze the recommended promotion time period for the product's corresponding video advertisement, and provide prompts.

[0016] The beneficial effects of the present invention are: 1. The present invention provides an advertising design multimedia information system, which analyzes the duration, background music and video special effects of video advertisements corresponding to target users, generates video advertisements, selects appropriate delivery time for delivery, monitors the effect of advertising delivery, and conducts corresponding promotion based on the delivery effect, thereby realizing the intelligent production and delivery of video advertisements, improving the attractiveness of video advertisements and the public's acceptance, enhancing the audience's viewing experience, ensuring the promotion effect, exposure rate and conversion rate of video advertisements, and also reducing the promotion cost of advertisements.

[0017] 2. The present invention selects an appropriate duration of video advertisements according to the viewing rate of videos of different durations by the target group of advertisement delivery, thereby ensuring the suitability of the duration of video advertisements, increasing the user's patience in watching, reducing the bounce rate caused by lengthy content, and increasing the effect of advertisement delivery.

[0018] 3. The present invention analyzes the novelty of background music and video special effects and the acceptability of the target group, thereby improving the attractiveness of subsequent videos and the public's acceptance, and also reflects the creativity of the advertisement, greatly improving the audience's viewing experience, and ensuring the publicity effect of the advertisement.

[0019] 4. When the effect of video advertising is poor, the present invention selects appropriate promotion time periods based on the online status of the account's own fan group at different time periods and the online status of the user group in the short video platform, which can increase the traffic of the advertisement, thereby ensuring the promotion effect, exposure rate and conversion rate of the video advertisement, and also reducing the promotion cost of the advertisement. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0021] Figure 1 This is a schematic diagram of the system structure connection of the present invention. DETAILED DESCRIPTION

[0022] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0023] See also Figure 1 As shown, an advertisement design multimedia information system includes: an advertisement design unit, a publishing monitoring unit and a database.

[0024] The advertisement design unit includes a template analysis module and an advertisement generation module.

[0025] The template analysis module is used to obtain the element data corresponding to the product after the target user uploads the product material data, and analyze the template of the video advertisement corresponding to the product. The template includes the recommended duration, recommended background music and recommended video special effects template.

[0026] It should be noted that target users need to upload relevant product advertising materials, including product posters and product images, to the advertising design multimedia information system. The advertising design multimedia information system contains multiple copyrighted background music and video special effects templates for users to use.

[0027] In a specific embodiment, the template of the video advertisement corresponding to the analysis product is as follows: S1. A number of images are obtained from the product material data, and then image recognition technology is used to obtain each element in each image and the number of occurrences of each element. The number of occurrences of each element is then divided by the total number of occurrences of each element to obtain the weight of each element, thereby using each element and the weight of each element as element data.

[0028] It should be noted that elements include but are not limited to textures and symbols.

[0029] S2. Obtain element data corresponding to each historical video advertisement from the database, and thereby analyze the target group data corresponding to each element.

[0030] Preferably, the target group data corresponding to each element is analyzed, and the specific process is as follows: the weight of each element corresponding to each historical video advertisement and the attraction index of each group are obtained from the database, and the attraction index set of each group corresponding to each weight in each element is statistically calculated.

[0031] It should be noted that the attraction index is the ratio of the number of users in each group to the total number of users in each group.

[0032] Extract the weight of each element in the advertisement, and thereby obtain the attraction index set of each group corresponding to the weight of each element; perform weighted average calculation on each attraction index in the attraction index set to obtain the average attraction index, which is used as the attraction index of each group corresponding to the weight of each element, and the attraction index of each group corresponding to the weight of each element is used as the target group data.

[0033] It should be noted that the groups include young women, young men and middle-aged women, among which young people can be aged 18-35 and middle-aged women can be aged 36-50.

[0034] S3. Obtain viewing data corresponding to each group from the advertising publishing platform, and then use the target group data corresponding to each element to analyze the recommended duration of the product's corresponding video advertisement.

[0035] The viewing data includes the total viewing time of each video watched by each user, the viewing time, and the weight of each element.

[0036] Preferably, the analysis product corresponds to the recommended duration of the video advertisement, and the specific process is as follows: the total duration of each video watched by each user, the watched duration and the weight of each element are obtained from the viewing data corresponding to each group.

[0037] The weights of each element in each video watched by each user in each group are compared with the weights of each element in the advertisement, and the overlap rate of elements between each video watched by each user in each group and the advertisement is analyzed. Then, the reference viewing videos corresponding to each user in each group are screened out.

[0038] In the above, the weight of each element in each watched video corresponding to each user in each group is divided by the weight of the corresponding element in the advertisement, and then the average is calculated to obtain the element overlap rate of each watched video and advertisement corresponding to each user in each group, and compared with the preset overlap rate threshold. The watched videos with a rate greater than the overlap rate threshold are used as reference watched videos.

[0039] It should be noted that the overlap rate threshold is the minimum critical value of the overlap rate, which is set by professionals and can be 0.7 or 0.8.

[0040] Divide the viewing time of each reference video by the total viewing time of each user in each group to obtain the viewing rate of each reference video for each user in each group. Integrate the reference videos with the same total viewing time to obtain the viewing rate of each user in each group for each total viewing time. Then calculate the average to obtain the average viewing rate of each total viewing time for each group.

[0041] Obtain the attraction index of each group corresponding to the weight of each element, multiply the attraction index of each group corresponding to the weight of each element, and then add them up to obtain the attraction index of each group.

[0042] The average viewing rate of videos watched for each total viewing time in each group is recorded as G xy , the attraction index of each group is recorded as F x , where x represents the number of each group, y represents the number corresponding to each total duration, and both x and y are positive integers.

[0043] Using the calculation formula Get the viewing attraction of the yth total viewing time of the video in the xth group Where n represents the total number of groups, and n is a positive integer.

[0044] The viewing attraction of each video with a total viewing duration in each group is added up to obtain the total viewing attraction of each video with a total viewing duration. The total duration corresponding to the maximum total viewing attraction is selected as the recommended duration of the video advertisement corresponding to the product.

[0045] The present invention selects an appropriate duration of video advertisements according to the viewing rate of videos of various durations by the target group of advertisement delivery, thereby ensuring the suitability of the duration of video advertisements, increasing the viewing patience of users, reducing the bounce rate caused by lengthy content, and increasing the effect of advertisement delivery.

[0046] S4. Obtain usage data and effect data corresponding to each background music from the advertising publishing platform, and use the target group data corresponding to each element to analyze the recommended background music for the product's corresponding video advertisement.

[0047] Among them, usage data includes release time, number of users in each group, number of times historical advertisements are used, and weight of each element in each historical advertisement; effect data includes user behavior data of each video; behavior data includes likes, reposts, and collections, etc.

[0048] It should be noted that the release time is the earliest time the background music was used.

[0049] Preferably, the analysis process of the recommended background music for the video advertisement corresponding to the product is as follows: obtaining the release time, the number of users in each group, the number of historical advertisement uses and the weight of each element in each historical advertisement from the usage data corresponding to each background music; obtaining the user behavior data of each background music corresponding to each video from the background music effect data.

[0050] The weights of each element in each historical advertisement in each background music are compared with the weights of each element in advertisements, and the reference advertisements in each background music are screened out. At the same time, the release time of each background music, the number of times it is used in historical advertisements and the number of reference advertisements are obtained to analyze the novelty of each background music.

[0051] For example, the novelty of each background music is analyzed in the following specific process: based on the release time of each background music, the release duration of each background music is obtained, which is recorded as T b At the same time, the number of historical advertisements used in each background music and the number of reference advertisements are recorded as M1 b and M2 b .

[0052] Using analytical formulas Get the novelty of the bth background music Where M1 represents the total number of historical advertising usage times, T represents the average length of time published, e represents a natural constant, γ1 and γ2 represent the proportional coefficient of advertising usage rate and the proportional coefficient of reference advertising usage rate, respectively.

[0053] It should be noted that the proportional coefficient of advertising usage rate and the proportional coefficient of reference advertising usage rate are set by professionals based on the importance of advertising usage rate and reference advertising usage rate. When the advertising usage rate has a greater impact on novelty, the proportional coefficient of advertising usage rate can be set to 0.6 and the proportional coefficient of reference advertising usage rate can be set to 0.4.

[0054] Short video platforms have a huge user base. When a hot topic or meme appears, a large number of content creators will quickly imitate and create similar content, resulting in an increase in content overlap. At the same time, the number of times the background music used in the video appears increases. Analyzing the novelty of background music can screen out background music with less spread, thereby increasing the freshness of video ads and increasing their attractiveness.

[0055] Using the user behavior data of each group corresponding to each background music and the user behavior data of each video corresponding to each background music, we analyze the acceptance of each group in each background music, which is recorded as J bx , where b represents the number of each background music, and b is a positive integer.

[0056] For example, the specific process of analyzing the acceptance of each group in each background music is as follows: based on the user behavior data of each video corresponding to each background music and each reference advertisement, the user behavior data of each reference advertisement video in each background music and the user behavior data of each ordinary video are obtained, and then the behavior data of each group corresponding to each reference advertisement video and each ordinary video are obtained from the user behavior data of each reference advertisement video in each background music and the user behavior data of each ordinary video, and weighted calculations are performed respectively to obtain the average behavior data of each group corresponding to the reference advertisement video and the ordinary video in each background music, which are respectively recorded as W1 bx and W2 bx .

[0057] Using analytical formulas Get the advertising radiation effect coefficient δ of the xth group in the bth background music bx , where ΔW represents the difference in reference behavior data.

[0058] It should be noted that the reference behavior data difference represents the maximum permitted behavior data difference, which is set by professionals. The reference behavior data difference can be 100 or 200. When the behavior data difference is greater than the reference behavior data difference, it indicates that users of the short video platform are much more receptive to ordinary videos with the same background music than to video advertisements, that is, the popularity of video advertisements in the background music is much lower than that of ordinary videos. The background music is not suitable for use in video advertisements and cannot increase the attractiveness of video advertisements.

[0059] The user behavior data of each group corresponding to each background music is recorded as W bx , substitute into the analytical formula Get the acceptance J of the xth group in the bth background music bx , where μ1 and μ2 represent the proportion coefficient of using user behavior data and the proportion coefficient of advertising radiation effect coefficient, respectively, W x Represents the sum of user behavior data of the x-th group.

[0060] It should be noted that the process of setting the proportional coefficient using user behavior data and the proportional coefficient of the advertising radiation effect coefficient is the same as the process of setting the proportional coefficient of the advertising usage rate and the proportional coefficient of the reference advertising usage rate, and will not be repeated here.

[0061] The larger the ratio of a group's user behavior data to the total user behavior data of the group in a certain background music, the more likely the group is to accept the background music. Choosing background music with a larger ratio can effectively increase the promotional effect of video advertising.

[0062] According to the analytical formula Get the attractiveness φ of the xth group in the bth background music bx ; Accumulate the attractiveness of each group in each background music to obtain the group attractiveness of each background music.

[0063] The novelty and attractiveness of each background music are recorded as and φ b , substitute into the calculation formula Get the priority eigenvalue α of the bth background music b , where ε1 and ε2 represent the proportional coefficients of novelty and attractiveness, respectively.

[0064] It should be noted that the process of setting the proportional coefficient of novelty and the proportional coefficient of attractiveness is the same as the process of setting the proportional coefficient of advertising usage rate and the proportional coefficient of reference advertising usage rate, and will not be repeated here.

[0065] The background music corresponding to the maximum priority feature value is selected from the priority feature values ​​of each background music as the recommended background music for the video advertisement corresponding to the product.

[0066] S5. Obtain usage data and effect data corresponding to each video special effects template from the advertising publishing platform, and use the target group data corresponding to each element to analyze the recommended video special effects template for the product's corresponding video advertisement.

[0067] It should be noted that the analysis process of the recommended video special effects template for the product corresponding video advertisement is the same as the analysis process of the recommended background music for the product corresponding video advertisement, and will not be repeated here.

[0068] The present invention analyzes the novelty of background music and video special effects and the acceptability of the target group, thereby improving the appeal of subsequent videos and the public's acceptance, and also reflects the creativity of the advertisement, greatly improving the audience's viewing experience, and ensuring the publicity effect of the advertisement.

[0069] The advertisement generation module is used to import product material data into the template of the product's corresponding video advertisement to generate the product's corresponding video advertisement.

[0070] The publishing monitoring unit includes an advertisement publishing module, a publishing effect monitoring module and an advertisement promotion analysis module.

[0071] The advertising publishing module is used to obtain the historical video publishing data corresponding to the target user account, and obtain the target group data corresponding to the product, analyze the recommended sending time of the product's video advertisement, and provide prompts.

[0072] The target user account corresponding to each historical video release data is obtained from the database. Each historical video release data includes the release time of each historical video, the behavior data of each group and the growth rate of the behavior data of each group.

[0073] In a specific embodiment, the recommended sending time of the video advertisement of the analyzed product is specifically processed as follows: the release time of each historical video, the behavior data of each group and the growth rate of the behavior data of each group are obtained from the historical video release data corresponding to the target user account, and the weight of each element corresponding to the product is extracted to obtain the attraction index of each group, thereby confirming the reference sending time set corresponding to the target user account.

[0074] Preferably, the release times of each historical video are compared with each other, and the historical videos corresponding to each release time are statistically obtained, and the behavioral data of each group in each historical video corresponding to each release time and the behavioral data growth rate of each group are normalized, and then the mean is calculated to obtain the publicity effect value of each group in each historical video corresponding to each release time, and then the publicity effect value of each group in each historical video corresponding to each release time is averaged to obtain the average publicity effect value of each group corresponding to each release time; the attraction index of each group is obtained, and the average publicity effect value of each group corresponding to each release time is multiplied by the attraction index of the corresponding group, and then added up to obtain the publicity feature value of each release time, and then compared with the preset publicity feature value threshold, and each release time with a publicity feature value greater than the publicity feature value threshold is selected as the reference sending time corresponding to the target user account, thereby obtaining the reference sending time set corresponding to the target user account.

[0075] Among them, the publicity feature value threshold is the critical value for judging whether the publicity effect is good or not, which is set by professionals. When the publicity feature value is greater than the publicity feature value threshold, it indicates that the publicity effect is good, otherwise it indicates that the publicity effect is not good. Suppose: the publicity feature value of a certain release time is 12, and the publicity feature value threshold is 18. 12<18, it indicates that the publicity effect at the release time is not good, and the video in the target user account is not suitable for release at that time.

[0076] Obtain behavioral data of each user in each group at each time period from the advertising publishing platform to confirm the reference sending time periods for the product's target groups.

[0077] Preferably, the behavioral data of each user in each group in each time period are accumulated to obtain the behavioral data of each group in each time period, the behavioral data of each group in each time period are normalized and multiplied with the attraction index of each group, and then accumulated to obtain the behavioral characteristic value of each time period, and then compared with the preset behavioral characteristic value threshold, and each time period with a behavioral characteristic value greater than the behavioral characteristic value threshold is selected as the reference sending time period.

[0078] Among them, the behavioral characteristic value threshold is the critical value for judging whether the user behavior is active, which is set by professionals. When the behavioral characteristic value is greater than the behavioral characteristic value threshold, it indicates that the user is relatively active, otherwise it indicates that the user is inactive. For example: the behavioral characteristic value of a certain period is 6, and the behavioral characteristic value threshold is 3. 6>3, indicating that the user is relatively active in this period, and video advertisements can be released during this period.

[0079] The recommended sending time of the product's video advertisement is determined based on the reference sending time set corresponding to the target user account and the reference sending time periods of the product's target group.

[0080] Among them, when a reference sending time in the reference sending time set corresponding to the target user account is within a reference sending period, the reference sending time is used as the candidate time, thereby obtaining various candidate times, and then selecting the smallest time as the recommended sending time for the product video advertisement.

[0081] The release effect monitoring module is used to arrange each monitoring moment according to the preset time interval after the product video advertisement is sent, collect the promotional data corresponding to each monitoring moment, and analyze the product advertising release effect.

[0082] It should be noted that promotional data includes page views and product purchases. Product video ads are linked to sales links, allowing users to purchase the product directly by clicking on the sales links. The preset time interval can be 30 minutes or 40 minutes, for example.

[0083] In a specific embodiment, the analysis effect of the advertising release effect of the product is as follows: the promotional data corresponding to each monitoring moment is calculated through the change rate calculation formula, the promotional data growth rate is analyzed, and the video advertising release time of the product is extracted, and the promotional data and promotional data growth rate at each monitoring moment in each historical video with the same release time are extracted from the database, and then a weighted average calculation is performed to obtain the reference promotional data and reference promotional data growth rate at each monitoring moment, and the data are input into the release effect analysis model to output the numerical value of the product's advertising release effect.

[0084] The value of the product's advertising effect includes values ​​1 and 0. When the value of the product's advertising effect is 1, it indicates that the product's advertising effect is good, otherwise it indicates that the product's advertising effect is poor.

[0085] For example, the promotional data corresponding to each monitoring moment is recorded as XC t , represents the number of each monitoring moment, t is a positive integer greater than 1, and the change rate calculation formula is: Get the promotional data growth rate BC, XC t-1 It represents the publicity data corresponding to the t-1th monitoring moment, and q represents the number of monitoring moments.

[0086] The reference publicity data and the reference publicity data growth rate at each monitoring moment are respectively recorded as XC t ′ and BC′, the expression of the release effect analysis model is Where ψ is the value of the product's advertising effect.

[0087] The advertising promotion analysis module is used to utilize the advertising release effect of the product, analyze the recommended promotion time period for the product's corresponding video advertisement, and provide prompts.

[0088] In a specific embodiment, the analysis of the recommended promotion period of the product corresponding to the video advertisement is as follows: when the product's advertising effect is good, it will not be promoted; when the product's advertising effect is poor, an advertisement recommendation will be made.

[0089] Obtain the number of online users in each group in each time period and the number of online users in each group in the target user account from the advertising publishing platform, analyze the recommendation index of each time period, and select the time period with the largest recommendation index as the recommended promotion period for the product's corresponding video advertisement.

[0090] Among them, the attraction index of each group is obtained, and the number of online users in each group in each time period and the number of online users of each group in the target user account are multiplied by the attraction index of the corresponding group respectively to obtain the recommendation index of each group in each time period and the recommendation index of each group in the target user account. Then, the recommendation index of each group in each time period and the recommendation index of each group in the target user account are added together to obtain the recommendation index of each time period.

[0091] The database is used to store the element data corresponding to each historical video advertisement, the weight of each element corresponding to each historical video advertisement, the attraction index of each group, the target user account corresponding to each historical video release data, and the publicity data and publicity data growth rate at each monitoring moment in each historical video.

[0092] When the effect of video advertising is poor, the present invention selects appropriate promotion time periods based on the online status of the account's own fan group at different time periods and the online status of the user group on the short video platform, which can increase the advertising traffic, thereby ensuring the promotion effect, exposure rate and conversion rate of the video advertising, and also reducing the promotion cost of the advertising.

[0093] The embodiment of the present invention generates video advertisements by analyzing the duration, background music and video special effects of video advertisements corresponding to target users, selects an appropriate delivery time for delivery, monitors the effect of advertisement delivery, and conducts corresponding promotion based on the delivery effect, thereby realizing intelligent production and delivery of video advertisements, improving the attractiveness of video advertisements and the public's acceptance, enhancing the audience's viewing experience, ensuring the promotion effect, exposure rate and conversion rate of video advertisements, and also reducing the promotion cost of advertisements.

[0094] The above content is merely an example and explanation of the concept of the present invention. Those skilled in the art may make various modifications or additions to the described specific embodiments or replace them in a similar manner. As long as they do not deviate from the concept of the invention or exceed the scope defined in this specification, they should all fall within the scope of protection of the present invention.

Claims

1. An advertisement design multimedia information system, characterized in that: include: Advertisement design unit and release monitoring unit; The advertisement design unit includes a template analysis module and an advertisement generation module; The template analysis module is used to obtain the element data corresponding to the product after the target user uploads the product material data, and analyze the template of the video advertisement corresponding to the product. The template includes the recommended duration, recommended background music and recommended video special effects template; The advertisement generation module is used to import product material data into the template of the product's corresponding video advertisement to generate the product's corresponding video advertisement; The publishing monitoring unit includes an advertisement publishing module, a publishing effect monitoring module and an advertisement promotion analysis module; The advertising publishing module is used to obtain the historical video publishing data corresponding to the target user account, obtain the target group data corresponding to the product, analyze the recommended sending time of the product's video advertisement, and provide prompts; The publishing effect monitoring module is used to arrange each monitoring moment according to the preset time interval after the product video advertisement is sent, collect the promotional data corresponding to each monitoring moment, and analyze the product advertising publishing effect; The advertising promotion analysis module is used to analyze the recommended promotion period for the product's corresponding video advertisement based on the product's advertising effect and provide prompts; The analysis product corresponds to the video ad template, and the specific process is as follows: S1. Obtain several images from product material data, then use image recognition technology to obtain each element in each image and the number of occurrences of each element. Then, divide the number of occurrences of each element by the total number of occurrences of all elements to obtain the weight of each element. Thus, each element and its weight are used as element data; S2. Obtaining element data corresponding to each historical video advertisement from the database, and analyzing target group data corresponding to each element; S3. Obtain viewing data corresponding to each group from the advertising publishing platform, and then use the target group data corresponding to each element to analyze the recommended duration of the video advertisement corresponding to the product; S4. Obtain usage data and effect data corresponding to each background music from the advertising publishing platform and use the target group data corresponding to each element to analyze the recommended background music for the product's corresponding video advertisement; S5. Obtain usage and performance data corresponding to each video special effects template from the advertising publishing platform, and use the target group data corresponding to each element to analyze the recommended video special effects template for the product's corresponding video advertisement; The specific process of analyzing the target group data corresponding to each element is as follows: Obtain the weights of each element corresponding to each historical video advertisement and the attraction index of each group from the database, and thereby calculate the attraction index set of each group corresponding to each weight in each element; Extract the weight of each element in the advertisement, thereby obtaining a set of attraction indexes corresponding to each group of weights of each element; perform weighted average calculation on each attraction index in the attraction index set to obtain an average attraction index, which is used as the attraction index corresponding to each group of weights of each element, and the attraction index corresponding to each group of weights of each element is used as the target group data; The recommended duration of the video advertisement corresponding to the analysis product is as follows: Obtain the total viewing time, viewing time, and weight of each element of each video for each user from the viewing data corresponding to each group; Compare the weights of each element in each video watched by each user in each group with the weights of each element in the advertisement, analyze the element overlap rate between each video watched by each user in each group and the advertisement, and then select the reference viewing videos corresponding to each user in each group; Divide the viewing time of each reference video by the total viewing time of each user in each group to obtain the viewing rate of each reference video in each group. Integrate the reference videos with the same total viewing time to obtain the viewing rate of each user in each group for each total viewing time. Then calculate the average to obtain the average viewing rate of each total viewing time in each group. Obtain the attraction index of each group corresponding to the weight of each element, multiply the attraction index of each group corresponding to the weight of each element, and then add them up to obtain the attraction index of each group; The average viewing rate of videos watched for each total viewing time in each group is recorded as G xy , the attraction index of each group is recorded as F x , where x represents the number of each group, y represents the number corresponding to each total duration, and both x and y are positive integers; Using the calculation formula Get the viewing attraction of the yth total viewing time of the video in the xth group Where n represents the total number of groups, n is a positive integer; The viewing attraction of each video with a total viewing duration in each group is added up to obtain the total viewing attraction of each video with a total viewing duration. The total duration corresponding to the maximum total viewing attraction is selected as the recommended duration of the video advertisement corresponding to the product.

2. The advertising design multimedia information system according to claim 1, characterized in that: The analysis process of the recommended background music for the video advertisement corresponding to the product is as follows: Obtain the release time, the number of users in each group, the number of times a historical ad is used, and the weight of each element in each historical ad from the usage data corresponding to each background music; obtain the user behavior data of each video corresponding to each background music from the background music effect data; The weights of each element in each historical advertisement in each background music are compared with the weights of each element in the advertisements, and the reference advertisements in each background music are screened out. At the same time, the release time of each background music, the number of times it is used in historical advertisements, and the number of reference advertisements are obtained to analyze the novelty of each background music; Using the user behavior data of each group corresponding to each background music and the user behavior data of each video corresponding to each background music, we analyze the acceptance of each group in each background music, which is recorded as J bx , where b represents the number of each background music, and b is a positive integer; According to the analytical formula Get the attractiveness φ of the xth group in the bth background music bx ; Accumulate the attractiveness of each group in each background music to obtain the group attractiveness of each background music; The novelty and attractiveness of each background music are recorded as and φ b , substitute into the calculation formula Get the priority eigenvalue α of the bth background music b , where ε1 and ε2 represent the proportional coefficients of novelty and attractiveness, respectively; The background music corresponding to the maximum priority feature value is selected from the priority feature values ​​of each background music as the recommended background music for the video advertisement corresponding to the product.

3. The advertising design multimedia information system according to claim 2, characterized in that: The specific process of analyzing the novelty of each background music is as follows: Based on the release time of each background music, obtain the release duration of each background music, recorded as T b At the same time, the number of historical advertisements used in each background music and the number of reference advertisements are recorded as M1 b and M2 b ; Using analytical formulas Get the novelty of the bth background music Where M1 represents the total number of historical advertising usage times, T represents the average length of time published, e represents a natural constant, γ1 and γ2 represent the proportional coefficient of advertising usage rate and the proportional coefficient of reference advertising usage rate, respectively.

4. The advertisement design multimedia information system according to claim 2, characterized in that: The specific process of analyzing the acceptance of each group in each background music is as follows: Based on the user behavior data of each video corresponding to each background music and each reference advertisement, the user behavior data of each reference advertisement video in each background music and the user behavior data of each ordinary video are obtained. Then, the behavior data of each group corresponding to each reference advertisement video and each ordinary video are obtained from the user behavior data of each reference advertisement video in each background music and the user behavior data of each ordinary video. Weighted calculations are performed on each group to obtain the average behavior data of each group corresponding to the reference advertisement video and the ordinary video in each background music, which are recorded as W1 respectively. bx and W2 bx ; Using analytical formulas Get the advertising radiation effect coefficient δ of the xth group in the bth background music bx , where ΔW represents the reference behavior data difference; The user behavior data of each group corresponding to each background music is recorded as W bx , substitute into the analytical formula Get the acceptance J of the xth group in the bth background music bx , where μ1 and μ2 represent the proportion coefficient of using user behavior data and the proportion coefficient of advertising radiation effect coefficient, respectively, W x Represents the sum of user behavior data of the x-th group.

5. The advertisement design multimedia information system according to claim 1, characterized in that: The recommended sending time of the video advertisement of the analyzed product is as follows: Obtain the release time of each historical video, the behavior data of each group, and the growth rate of each group's behavior data from the target user account's corresponding historical video release data, and extract the weight of each element corresponding to the product and the attractiveness index of each group, thereby confirming the reference sending time set corresponding to the target user account; Obtain behavioral data of each user in each group at each time period from the advertising publishing platform to confirm the reference delivery time period for the product target group; The recommended sending time of the product's video advertisement is determined based on the reference sending time set corresponding to the target user account and the reference sending time periods of the product's target group.

6. The advertisement design multimedia information system according to claim 1, characterized in that: The analysis results of the advertising effect of the product are as follows: The promotional data corresponding to each monitoring moment is calculated through a rate of change formula to analyze the promotional data growth rate, and the product video advertisement release time is extracted. The promotional data and promotional data growth rate at each monitoring moment in each historical video with the same release time are extracted from the database, and then a weighted average calculation is performed to obtain the reference promotional data and reference promotional data growth rate at each monitoring moment. These are input into the release effect analysis model to output the numerical value of the product's advertising release effect; The value of the product's advertising effect includes values ​​1 and 0. When the value of the product's advertising effect is 1, it indicates that the product's advertising effect is good, otherwise it indicates that the product's advertising effect is poor.

7. The advertisement design multimedia information system according to claim 1, characterized in that: The analysis product corresponds to the recommended promotion period of the video advertisement. The specific process is as follows: When the product's advertising effect is good, it will not be promoted. When the product's advertising effect is poor, it will be recommended. Obtain the number of online users in each group in each time period and the number of online users in each group in the target user account from the advertising publishing platform, analyze the recommendation index of each time period, and select the time period with the largest recommendation index as the recommended promotion period for the product's corresponding video advertisement.

Images