Advertisement intelligent design system based on big data

Abstract

The invention discloses an intelligent advertisement design system based on big data, which relates to the technical field of data analysis, and comprises an acquisition module, a processing module, an analysis module and a design module, and is characterized in that associated materials are matched, a material group is obtained, the first duration of a first video is counted, a first ratio is calculated, a scoring model is constructed, a second duration is set, and a first operation is executed. And screening the key frame, setting a timestamp, setting a picture parameter and a sound parameter, and generating an advertisement fragment. According to the automatic advertisement design process, manual operation is reduced, time and resources are saved, related materials are matched according to product names, the attraction of advertisements is improved, the rhythm and duration of the advertisements are optimized, the watching experience of audiences is improved, a scoring model is constructed, advertisement design is more scientific and objective, key frames are intelligently screened, and the advertisement design efficiency is improved. And the picture parameters and the sound parameters are set according to the timestamps, so that different playing environments and audience requirements are met.

Description

Technical Field

[0001] The present invention relates to the technical field of data analysis, and in particular to an intelligent advertising design system based on big data. Background Art

[0002] In recent years, intelligent advertising design is in a stage of rapid development, the application of AI technology has continued to deepen, and significant progress has been made in advertising creativity, delivery efficiency, and personalized marketing. Machine learning can use data-driven methods to create advertising creativity and accurately identify the interests and preferences of the target audience by analyzing massive user data to achieve personalized advertising.

[0003] At present, in a Chinese invention patent with publication number CN 112508612B, a training advertising creative generation model, a method for generating advertising creatives and related devices are disclosed. By taking the user's query text and the corresponding click text as question-answer type training text, the first entity word in the query text, the first commodity category corresponding to the first entity word, and the first commodity feature of the commodity in the click text are identified; through the first entity word, the first commodity category, the first commodity feature and the question-answer type training text, a preset text generation network is trained to obtain an advertising creative generation model, and a question-answer type advertising creative is generated. However, the related technology does not optimize the advertisement according to the relevant parameters of the product and the audience's response data, which is not conducive to the pertinence and scientific nature of the advertising delivery. No reference node is set to adjust the relevant parameters such as sound and picture, which easily causes the sound to be too fast or delayed, which is not conducive to the precision and accuracy of the advertisement. Summary of the invention

[0004] The technical problem solved by the present invention is that the related technology does not optimize the advertisement according to the relevant parameters of the product and the audience's response data, which is not conducive to the pertinence and scientific nature of the advertisement placement. No reference node is set to adjust the relevant parameters such as sound and picture, which easily causes the sound to be too fast or delayed, which is not conducive to the precision and accuracy of the advertisement.

[0005] In order to solve the above technical problems, the present invention provides the following technical solutions: In the first aspect, an intelligent advertising design system based on big data includes a collection module, a processing module, an analysis module and a design module;

[0006] The acquisition module obtains the product name and matches the associated material according to the product name;

[0007] The processing module sets the appearance order of the associated materials and the duration of the advertisement, obtains a material group according to the associated materials, and first numbers the material groups according to the appearance order, automatically generates a first video according to the material groups after the first numbering, and counts the first duration of the first video;

[0008] The analysis module obtains historical advertising data and corresponding historical audience scores, calculates a first ratio according to the historical advertising data, performs a multivariate regression analysis on the first ratio and the historical audience scores, constructs a scoring model, sets a first ratio of the current material group according to the scoring model, calculates the appearance time of the current material group according to the first ratio of the current material group, records it as a second duration, and performs a first operation according to the second duration and the first duration;

[0009] In response to the first operation, the design module preprocesses the first video to obtain the first image, acquires the product image, calculates the correlation between the first image and the product image, filters the first image to be deleted according to the correlation, filters the key frames according to the correlation of the remaining first images, sets the timestamp according to the key frames, sets the picture parameters and the sound parameters according to the timestamp, and generates an advertising clip.

[0010] As a preferred solution of the big data-based advertising intelligent design system described in the present invention, the acquisition module obtains the product name, calls the AI ​​image software, inputs the product name into the AI ​​image software, matches the associated material corresponding to the product name, obtains the product category described by the product name, and stores the product name and the corresponding associated material in the category of the product category in the advertising database;

[0011] The associated materials include product brand, product origin, product raw materials, product characteristics, product functions, product advantages and background elements. The background elements include dynamic background and static background. The background elements are acquired through historical advertising experience. The static background includes static plants, static animals, static people and static natural landscapes. The dynamic background includes dynamic plants, dynamic animals, dynamic people and dynamic natural landscapes.

[0012] As a preferred solution of the big data-based advertising intelligent design system described in the present invention, wherein: the processing module sets the appearance order of the associated materials and the advertising duration, the associated materials can be combined and appear at the same time, and the advertising duration is set to the current advertising duration;

[0013] The associated materials with the same appearance order are combined to obtain a material group, and the material group is first numbered according to the appearance order, and the first number is represented by X i , where i is a natural number, denoted as the current material group;

[0014] Call up AI video software, input the material group after the first number into the AI ​​video software, automatically generate a first video, and count the length of the first video, which is recorded as the first length.

[0015] As a preferred solution of the big data-based advertising intelligent design system described in the present invention, wherein: the analysis module obtains the product category to which the product name belongs, retrieves the advertising database, inputs the product category into the advertising database, and matches the historical advertising data corresponding to the product category and the corresponding historical audience rating;

[0016] The historical advertisement data includes a historical material group, a material appearance time of the historical material group, and a total time of historical advertisements. The material group is compared with the historical material group. When the material group is the same as the historical material group, the historical material group is set to a corresponding first number. When the material group is different from the historical material group, the historical material group is set to a second number. The second number is represented by Y. j , where j is a natural number.

[0017] As a preferred solution of the big data-based advertising intelligent design system described in the present invention, the analysis module calculates the ratio of the appearance time of the historical material group to the total duration of the historical advertisements, recorded as the first ratio, and sets the first number or the second number corresponding to the historical material group as the label of the corresponding first ratio;

[0018] Performing a multivariate regression analysis on the first ratio and historical audience ratings using the first ratio, constructing a rating model, and setting the first ratio of the current material group according to the rating model;

[0019] The appearance duration of the current material group is calculated according to the first ratio of the current material group and recorded as the second duration. The first operation is performed according to the second duration and the first duration. The first operation includes not adjusting the first duration and adjusting the first duration.

[0020] As a preferred solution of the big data-based advertising intelligent design system described in the present invention, the calculation expression of the scoring model is:

[0021]

[0022] Where P is the historical audience rating, m1 and m2 represent the range of the second number, and m1 and m2 are natural numbers, n1 and n2 represent the range of the first number, and n1 and n2 are natural numbers, T i and T j are the first duration of the first numbered historical material group and the first duration of the second numbered historical material group, respectively. i and C j The coefficients of the first duration of the first numbered historical material group and the coefficients of the first duration of the second numbered historical material group, C i and C j is a constant;

[0023] The calculation logic of the second duration includes obtaining a coefficient of the first duration of the historical material group with the first number, recorded as the first coefficient, performing weighted calculation on the first coefficient to obtain a first sum value, calculating a ratio of the first coefficient to the first sum value, recorded as the second ratio, wherein the second ratio represents a proportion of the current material group to the current advertisement duration, and calculating a product of the current advertisement duration and the second ratio, recorded as the second duration;

[0024] The second duration is compared with the first duration, and when the second duration is greater than the first duration, the first operation is set to adjust the first duration; otherwise, the first operation is set to not adjust the first duration.

[0025] As a preferred solution of the big data-based advertising intelligent design system described in the present invention, when the first operation is to adjust the first duration, the design module pre-processes the first video in response to the first operation, and the pre-processing includes frame processing and windowing processing to obtain a first image;

[0026] Retrieving a product database, inputting a product name into the product database, matching a product image corresponding to the product name, calculating a correlation between the first image and the product image, and filtering the first image to be deleted according to the correlation until the first duration is equal to the second duration, and then stopping the deletion;

[0027] Sort the relevance of the remaining first images in descending order, set the first value as the relevance threshold, compare the relevance with the first value, and when the relevance is greater than or equal to the first value, set the first image as a key frame, otherwise, end the screening;

[0028] A time point corresponding to a key frame is obtained, where the time point is represented as a time point at which the key frame is located in a first image, a timestamp is set at the time point corresponding to the key frame, and picture parameters and sound parameters are set according to the timestamp, where the picture-related parameters include frame size, color saturation, frame rate, resolution, and picture aspect ratio, and the sound-related parameters include music name, vocal segment, music volume, vocal volume, vocal tint, vocal speed, music speed, music bit depth, and vocal bit depth, and an advertising segment is generated.

[0029] As a preferred solution of the big data-based advertising intelligent design system described in the present invention, the calculation logic of the relevance includes:

[0030] Obtain any first image, extract text feature quantity and shape feature quantity of the first image, and record them as first feature quantity and second feature quantity;

[0031] Obtain product brand, product origin, product raw materials, product characteristics, product functions and product advantages, retrieve product database, obtain product packaging pictures, product raw material pictures and product brand pictures, extract shape feature quantities of product packaging pictures, shape feature quantities of product raw material pictures and shape feature quantities of product brand pictures, record them as third feature quantities, extract text feature quantities of product origin, text feature quantities of product characteristics, text feature quantities of product functions and text feature quantities of product advantages, record them as fourth feature quantities, calculate the similarity between the first feature quantity and the fourth feature quantity by cosine similarity formula, record them as first similarity, calculate the similarity between the second feature quantity and the third feature quantity by cosine similarity formula, record them as second similarity, calculate the sum of the first similarity and the second similarity, record them as correlation;

[0032] The setting logic of the picture parameters and sound parameters includes:

[0033] According to the time sequence, the first timestamp is obtained and recorded as the starting timestamp, the current material group at the starting timestamp is identified, the AI ​​editing software is called, the current material group is input into the AI ​​editing software, and the sound parameters and picture parameters are automatically matched.

[0034] In a second aspect, the present invention provides an electronic device comprising a memory, a processor and storage in the memory, wherein the memory stores computer-readable instructions, and when the computer-readable instructions are executed by the processor, the steps in any one of the systems described above are executed.

[0035] In a third aspect, the present invention provides a storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the steps in the system described in any one of the above items are executed.

[0036] The beneficial effects of the present invention are as follows: the automated advertising design process reduces manual operations, saves time and resources, matches related materials according to product names, makes advertising content more in line with product characteristics, and enhances the attractiveness of advertisements; optimizes the rhythm and duration of advertisements by processing the appearance order and advertisement duration set by the processing module, and improves the audience's viewing experience; uses historical advertising data and audience ratings to perform multivariate regression analysis and build a rating model, making advertising design more scientific and objective; through preprocessing of the first video and calculation of the correlation of the product image, intelligently filters key frames to improve the accuracy of editing; through calculating the correlation between the first image and the product image, filters out images that best represent the product characteristics and enhances the visual appeal of the advertisement; sets picture parameters and sound parameters according to timestamps, and dynamically adjusts the generated advertising clips to adapt to different playback environments and audience needs. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1A basic flow chart of an advertising intelligent design system based on big data is provided for one embodiment of the present invention. DETAILED DESCRIPTION

[0038] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings. It is obvious that the described embodiments are only part of the embodiments of the present invention, but not all of the embodiments.

[0039] Example, see Figure 1 , is an embodiment of the present invention, which provides an advertising intelligent design system based on big data, including a collection module, a processing module, an analysis module and a design module;

[0040] The acquisition module obtains the product name and matches the associated material according to the product name;

[0041] The processing module sets the appearance order of the associated materials and the duration of the advertisement, obtains a material group according to the associated materials, and first numbers the material groups according to the appearance order, automatically generates a first video according to the material groups after the first numbering, and counts the first duration of the first video;

[0042] The analysis module obtains historical advertising data and corresponding historical audience scores, calculates a first ratio according to the historical advertising data, performs a multivariate regression analysis on the first ratio and the historical audience scores, constructs a scoring model, sets a first ratio of the current material group according to the scoring model, calculates the appearance time of the current material group according to the first ratio of the current material group, records it as a second duration, and performs a first operation according to the second duration and the first duration;

[0043] In response to the first operation, the design module preprocesses the first video to obtain the first image, acquires the product image, calculates the correlation between the first image and the product image, filters the first image to be deleted according to the correlation, filters the key frames according to the correlation of the remaining first images, sets the timestamp according to the key frames, sets the picture parameters and the sound parameters according to the timestamp, and generates an advertising clip.

[0044] The automated advertisement design process of the present invention reduces manual operations, saves time and resources, matches related materials according to product names, makes advertisement content more in line with product characteristics, and enhances advertisement appeal; optimizes advertisement rhythm and duration by processing the appearance order and advertisement duration set by the processing module, and enhances audience viewing experience; conducts multivariate regression analysis using historical advertisement data and audience ratings, and constructs a rating model, making advertisement design more scientific and objective; intelligently filters key frames by preprocessing the first video and calculating the correlation of product images, and improves editing accuracy; filters out images that best represent product characteristics by calculating the correlation between the first image and the product image, and enhances the visual appeal of the advertisement; sets picture parameters and sound parameters according to timestamps, and dynamically adjusts the generated advertisement clips to adapt to different playback environments and audience needs.

[0045] The acquisition module obtains the product name, calls the AI ​​image software, inputs the product name into the AI ​​image software, matches the associated material corresponding to the product name, obtains the product category described by the product name, and stores the product name and the corresponding associated material in the category of the product category in the advertisement database;

[0046] The associated materials include product brand, product origin, product raw materials, product characteristics, product functions, product advantages and background elements. The background elements include dynamic background and static background. The background elements are acquired through historical advertising experience. The static background includes static plants, static animals, static people and static natural landscapes. The dynamic background includes dynamic plants, dynamic animals, dynamic people and dynamic natural landscapes.

[0047] In the specific implementation, AI image software is used to automatically match related materials according to product names, which improves the efficiency of material acquisition, reduces manual intervention, and obtains a variety of product-related materials, including product brand, origin, raw materials, characteristics, functions, advantages and background elements, making the advertising design more comprehensive and attractive. Background elements include dynamic and static backgrounds, which can be selected based on historical advertising experience, enhancing the visual effect and appeal of the advertisement. By storing product names and corresponding related materials in the advertising database, the system can provide data support for future advertising design, facilitate analysis and optimization, and simplify the advertising design process and improve overall work efficiency through automated material matching and classification.

[0048] The processing module sets the appearance order of the associated materials and the duration of the advertisement, the associated materials can be combined and appear at the same time, and the duration of the advertisement is set to the current duration of the advertisement;

[0049] Combining related materials with the same appearance order to obtain a material group, and assigning a first number to the material group according to the appearance order, wherein the first number is represented by Xi, where i is a natural number and is recorded as the current material group;

[0050] Call up AI video software, input the material group after the first number into the AI ​​video software, automatically generate a first video, and count the length of the first video, which is recorded as the first length.

[0051] The analysis module obtains the product category to which the product name belongs, retrieves an advertisement database, inputs the product category into the advertisement database, and matches historical advertisement data corresponding to the product category and corresponding historical audience ratings;

[0052] The historical advertising data includes historical material groups, material appearance duration of historical material groups and total duration of historical advertisements. The material group is compared with the historical material group. When the material group is the same as the historical material group, the historical material group is set to a corresponding first number. When the material group is different from the historical material group, the historical material group is set to a second number. The second number is represented by Yj, where j is a natural number.

[0053] In specific implementation, setting the advertisement duration to the current advertisement duration helps to maintain the compactness and appeal of the advertisement and avoid audience loss due to overly long advertisements. It allows the combination of related materials with the same order of appearance, which increases the flexibility and diversity of advertisement creation. The material groups are systematically organized through the first number (Xi), which facilitates the tracking and management of advertising materials, improves the logic and organization of material use, and the automated video generation and material management reduce the dependence on professional video producers, thereby reducing the cost of advertisement production.

[0054] The analysis module calculates the ratio of the appearance duration of the historical material group to the total duration of the historical advertisements, records it as a first ratio, and sets the first number or the second number corresponding to the historical material group as a label of the corresponding first ratio;

[0055] Performing a multivariate regression analysis on the first ratio and historical audience ratings using the first ratio, constructing a rating model, and setting the first ratio of the current material group according to the rating model;

[0056] The appearance duration of the current material group is calculated according to the first ratio of the current material group and recorded as the second duration. The first operation is performed according to the second duration and the first duration. The first operation includes not adjusting the first duration and adjusting the first duration.

[0057] The calculation expression of the scoring model is:

[0058]

[0059] Where P is the historical audience rating, m1 and m2 represent the range of the second number, and m1 and m2 are natural numbers, n1 and n2 represent the range of the first number, and n1 and n2 are natural numbers, T i and T j are the first duration of the first numbered historical material group and the first duration of the second numbered historical material group, respectively. i and C j The coefficients of the first duration of the first numbered historical material group and the coefficients of the first duration of the second numbered historical material group, C i and C j is a constant;

[0060] The calculation logic of the second duration includes obtaining a coefficient of the first duration of the historical material group with the first number, recorded as the first coefficient, performing weighted calculation on the first coefficient to obtain a first sum value, calculating a ratio of the first coefficient to the first sum value, recorded as the second ratio, wherein the second ratio represents a proportion of the current material group to the current advertisement duration, and calculating a product of the current advertisement duration and the second ratio, recorded as the second duration;

[0061] The second duration is compared with the first duration, and when the second duration is greater than the first duration, the first operation is set to adjust the first duration; otherwise, the first operation is set to not adjust the first duration.

[0062] In specific implementation, by calculating the ratio of the appearance time of the historical material group to the total duration of the historical advertisements (the first ratio), the efficiency and effect of the material use can be analyzed more accurately, and the first number or the second number of the historical material group can be set as the label of the corresponding first ratio, thereby realizing label management of the material, facilitating the tracking and optimization of the use of advertising materials, and constructing a scoring model using multivariate regression analysis to more scientifically predict and evaluate the performance of different material groups. The first ratio of the current material group is set according to the scoring model, and the appearance time (the second time) is calculated, which helps to optimize the rhythm and effect of the advertisement. The first operation is performed according to the second time and the first time, including not adjusting or adjusting the first time, thereby providing flexibility in adjusting the advertisement duration to adapt to different advertising strategies and market changes.

[0063] When the first operation is to adjust the first duration, the design module preprocesses the first video in response to the first operation, wherein the preprocessing includes frame processing and windowing processing to obtain a first image;

[0064] Retrieving a product database, inputting a product name into the product database, matching a product image corresponding to the product name, calculating a correlation between the first image and the product image, and filtering the first image to be deleted according to the correlation until the first duration is equal to the second duration, and then stopping the deletion;

[0065] Sort the relevance of the remaining first images in descending order, set the first value as the relevance threshold, compare the relevance with the first value, and when the relevance is greater than or equal to the first value, set the first image as a key frame, otherwise, end the screening;

[0066] A time point corresponding to a key frame is obtained, where the time point is represented as a time point at which the key frame is located in a first image, a timestamp is set at the time point corresponding to the key frame, and picture parameters and sound parameters are set according to the timestamp, where the picture-related parameters include frame size, color saturation, frame rate, resolution, and picture aspect ratio, and the sound-related parameters include music name, vocal segment, music volume, vocal volume, vocal tint, vocal speed, music speed, music bit depth, and vocal bit depth, and an advertising segment is generated.

[0067] In specific implementation, the first video is pre-processed through frame processing and windowing processing, which can effectively extract important image information in the video, lay the foundation for subsequent image analysis and processing, retrieve the product database and match the corresponding product image, ensure that the advertising content is highly relevant to the product characteristics, enhance the pertinence and attractiveness of the advertisement, and continuously screen the first image to be deleted before the first duration is equal to the second duration to ensure that the final generated advertising segment meets the duration requirements and enhance the rhythm of the advertisement. The time point corresponding to the key frame is obtained and the timestamp is set to ensure that the advertising segment can accurately display important content when playing and enhance the audience's viewing experience. The relevant parameters of the picture and sound are set according to the timestamp, including frame size, color saturation, volume, etc., which can improve the overall quality and viewing experience of the advertisement.

[0068] The calculation logic of the correlation includes:

[0069] Obtain any first image, extract text feature quantity and shape feature quantity of the first image, and record them as first feature quantity and second feature quantity;

[0070] Obtain product brand, product origin, product raw materials, product characteristics, product functions and product advantages, retrieve product database, obtain product packaging pictures, product raw material pictures and product brand pictures, extract shape feature quantities of product packaging pictures, shape feature quantities of product raw material pictures and shape feature quantities of product brand pictures, record them as third feature quantities, extract text feature quantities of product origin, text feature quantities of product characteristics, text feature quantities of product functions and text feature quantities of product advantages, record them as fourth feature quantities, calculate the similarity between the first feature quantity and the fourth feature quantity by cosine similarity formula, record them as first similarity, calculate the similarity between the second feature quantity and the third feature quantity by cosine similarity formula, record them as second similarity, calculate the sum of the first similarity and the second similarity, record them as correlation;

[0071] The setting logic of the picture parameters and sound parameters includes:

[0072] According to the time sequence, the first timestamp is obtained and recorded as the starting timestamp, the current material group at the starting timestamp is identified, the AI ​​editing software is called, the current material group is input into the AI ​​editing software, and the sound parameters and picture parameters are automatically matched.

[0073] In the specific implementation, by extracting the text feature and shape feature of the first image, as well as the text and shape feature of product-related information, the similarity between the image and the product can be fully captured, and the cosine similarity formula is used to calculate the similarity between the first feature and the fourth feature, and the second feature and the third feature, providing a quantitative method to evaluate the correlation between the image and the product. Through accurate correlation calculation, it is ensured that the advertising content is highly matched with the product, thereby improving the personalization and targeting of the advertisement. The calculation of the correlation is based on data and algorithms, making the selection of advertising materials more scientific and objective.

[0074] The automated advertisement design process of the present invention reduces manual operations, saves time and resources, matches related materials according to product names, makes advertisement content more in line with product characteristics, and enhances advertisement appeal; optimizes advertisement rhythm and duration by processing the appearance order and advertisement duration set by the processing module, and enhances audience viewing experience; conducts multivariate regression analysis using historical advertisement data and audience ratings, and constructs a rating model, making advertisement design more scientific and objective; intelligently filters key frames by preprocessing the first video and calculating the correlation of product images, and improves editing accuracy; filters out images that best represent product characteristics by calculating the correlation between the first image and the product image, and enhances the visual appeal of the advertisement; sets picture parameters and sound parameters according to timestamps, and dynamically adjusts the generated advertisement clips to adapt to different playback environments and audience needs.

[0075] It should be understood by those skilled in the art that the embodiments of the present invention can be provided as methods, systems or computer program products. Therefore, the present invention can be in the form of a complete hardware embodiment, a complete software embodiment or an embodiment combining software and hardware. Moreover, the present invention can be in the form of a computer program product implemented on one or more computer-usable storage media containing computer-usable program codes. Wherein, the storage medium is implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (Static Random Access Memory, referred to as SRAM), electrically erasable programmable read-only memory (Electrically Erasable Programmable Read-Only Memory, referred to as EEPROM), erasable programmable read-only memory (Erasable Programmable Read Only Memory, referred to as EPROM), programmable read-only memory (Programmable Red-Only Memory, referred to as PROM), read-only memory (Read-Only Memory, referred to as ROM), magnetic memory, flash memory, disk or optical disk. These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 Functions specified in one or more boxes

[0076] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions of the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention should be included in the scope of the claims of the present invention.

Claims

1. An intelligent advertising design system based on big data, characterized in that: It includes acquisition module, processing module, analysis module and design module; The acquisition module obtains the product name and matches the associated material according to the product name; The processing module sets the appearance order of the associated materials and the duration of the advertisement, obtains a material group according to the associated materials, and first numbers the material groups according to the appearance order, automatically generates a first video according to the material groups after the first numbering, and counts the first duration of the first video; The analysis module obtains historical advertising data and corresponding historical audience scores, calculates a first ratio according to the historical advertising data, performs a multivariate regression analysis on the first ratio and the historical audience scores, constructs a scoring model, sets a first ratio of the current material group according to the scoring model, calculates the appearance time of the current material group according to the first ratio of the current material group, records it as a second duration, and performs a first operation according to the second duration and the first duration; In response to the first operation, the design module preprocesses the first video to obtain the first image, acquires the product image, calculates the correlation between the first image and the product image, filters the first image to be deleted according to the correlation, filters the key frames according to the correlation of the remaining first images, sets the timestamp according to the key frames, sets the picture parameters and the sound parameters according to the timestamp, and generates an advertising clip.

2. The advertising intelligent design system based on big data as claimed in claim 1, characterized in that: The acquisition module obtains the product name, calls the AI ​​image software, inputs the product name into the AI ​​image software, matches the associated material corresponding to the product name, obtains the product category described by the product name, and stores the product name and the corresponding associated material in the category of the product category in the advertisement database; The associated materials include product brand, product origin, product raw materials, product characteristics, product functions, product advantages and background elements. The background elements include dynamic background and static background. The background elements are acquired through historical advertising experience. The static background includes static plants, static animals, static people and static natural landscapes. The dynamic background includes dynamic plants, dynamic animals, dynamic people and dynamic natural landscapes.

3. The advertising intelligent design system based on big data as claimed in claim 1, characterized in that: The processing module sets the appearance order of the associated materials and the duration of the advertisement, the associated materials can be combined and appear at the same time, and the duration of the advertisement is set to the current duration of the advertisement; The associated materials with the same appearance order are combined to obtain a material group, and the material group is first numbered according to the appearance order, and the first number is represented by X i , where i is a natural number, denoted as the current material group; Call up AI video software, input the material group after the first number into the AI ​​video software, automatically generate a first video, and count the length of the first video, which is recorded as the first length.

4. The advertising intelligent design system based on big data as claimed in claim 1, characterized in that: The analysis module obtains the product category to which the product name belongs, retrieves an advertisement database, inputs the product category into the advertisement database, and matches historical advertisement data corresponding to the product category and corresponding historical audience ratings; The historical advertisement data includes a historical material group, a material appearance time of the historical material group, and a total time of historical advertisements. The material group is compared with the historical material group. When the material group is the same as the historical material group, the historical material group is set to a corresponding first number. When the material group is different from the historical material group, the historical material group is set to a second number. The second number is represented by Y. j , where j is a natural number.

5. The big data-based advertising intelligent design system according to claim 4, characterized in that: The analysis module calculates the ratio of the appearance duration of the historical material group to the total duration of the historical advertisements, records it as a first ratio, and sets the first number or the second number corresponding to the historical material group as a label of the corresponding first ratio; Performing a multivariate regression analysis on the first ratio and historical audience ratings using the first ratio, constructing a rating model, and setting the first ratio of the current material group according to the rating model; The appearance duration of the current material group is calculated according to the first ratio of the current material group and recorded as the second duration. The first operation is performed according to the second duration and the first duration. The first operation includes not adjusting the first duration and adjusting the first duration.

6. The big data-based advertising intelligent design system according to claim 1, characterized in that: The calculation expression of the scoring model is: Where P is the historical audience rating, m1 and m2 represent the range of the second number, and m1 and m2 are natural numbers, n1 and n2 represent the range of the first number, and n1 and n2 are natural numbers, T i and T j are the first duration of the first numbered historical material group and the first duration of the second numbered historical material group, respectively. i and C j The coefficients of the first duration of the first numbered historical material group and the coefficients of the first duration of the second numbered historical material group, C i and C j is a constant; The calculation logic of the second duration includes obtaining a coefficient of the first duration of the historical material group with the first number, recorded as the first coefficient, performing weighted calculation on the first coefficient to obtain a first sum value, calculating a ratio of the first coefficient to the first sum value, recorded as the second ratio, wherein the second ratio represents a proportion of the current material group to the current advertisement duration, and calculating a product of the current advertisement duration and the second ratio, recorded as the second duration; The second duration is compared with the first duration, and when the second duration is greater than the first duration, the first operation is set to adjust the first duration; otherwise, the first operation is set to not adjust the first duration.

7. The advertising intelligent design system based on big data as claimed in claim 1, characterized in that: When the first operation is to adjust the first duration, the design module preprocesses the first video in response to the first operation, wherein the preprocessing includes frame processing and windowing processing to obtain a first image; Retrieving a product database, inputting a product name into the product database, matching a product image corresponding to the product name, calculating a correlation between the first image and the product image, and filtering the first image to be deleted according to the correlation until the first duration is equal to the second duration, and then stopping the deletion; Sort the relevance of the remaining first images in descending order, set the first value as the relevance threshold, compare the relevance with the first value, and when the relevance is greater than or equal to the first value, set the first image as a key frame, otherwise, end the screening; A time point corresponding to a key frame is obtained, where the time point is represented as a time point at which the key frame is located in a first image, a timestamp is set at the time point corresponding to the key frame, and picture parameters and sound parameters are set according to the timestamp, where the picture-related parameters include frame size, color saturation, frame rate, resolution, and picture aspect ratio, and the sound-related parameters include music name, vocal segment, music volume, vocal volume, vocal tint, vocal speed, music speed, music bit depth, and vocal bit depth, and an advertising segment is generated.

8. The big data-based advertising intelligent design system according to claim 7, characterized in that: The calculation logic of the correlation includes: Obtain any first image, extract text feature quantity and shape feature quantity of the first image, and record them as first feature quantity and second feature quantity; Obtain product brand, product origin, product raw materials, product characteristics, product functions and product advantages, retrieve product database, obtain product packaging pictures, product raw material pictures and product brand pictures, extract shape feature quantities of product packaging pictures, shape feature quantities of product raw material pictures and shape feature quantities of product brand pictures, record them as third feature quantities, extract text feature quantities of product origin, text feature quantities of product characteristics, text feature quantities of product functions and text feature quantities of product advantages, record them as fourth feature quantities, calculate the similarity between the first feature quantity and the fourth feature quantity by cosine similarity formula, record them as first similarity, calculate the similarity between the second feature quantity and the third feature quantity by cosine similarity formula, record them as second similarity, calculate the sum of the first similarity and the second similarity, record them as correlation; The setting logic of the picture parameters and sound parameters includes: According to the time sequence, the first timestamp is obtained and recorded as the starting timestamp, the current material group at the starting timestamp is identified, the AI ​​editing software is called, the current material group is input into the AI ​​editing software, and the sound parameters and picture parameters are automatically matched.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the advertising intelligent design system based on big data as described in any one of claims 1 to 8 is implemented.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by the processor, it implements the advertising intelligent design system based on big data as described in any one of claims 1-8.

Images