Advertisement delivery system and method based on data identification

Abstract

The invention discloses an advertisement putting system and method based on data recognition, and the system comprises a data collection module, a data processing module, a database module, and an advertisement putting adjustment module. The delivery method comprises the following steps: collecting user behavior data on a short video platform to form a behavior data set; preprocessing and screening the user behavior data; reading and storing to-be-delivered advertisements, and establishing a pushed advertisement database; reading a pushed advertisement form, and matching advertisement data in a pushed advertisement database; and dividing advertisement putting time periods and advertisement putting forms, and matching and pushing advertisement data in the advertisement database. According to the method and the device, different advertisement putting strategies can be adopted according to user classification types and user behavior types by dividing different time periods of the whole day; by reasonably planning the putting time and form of the advertisement, not only can the exposure of the advertisement be ensured, but also the interference to the watching experience of the user can be effectively reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of advertisement delivery, and in particular to an advertisement delivery system and method based on data identification. Background Art

[0002] Short video ads refer to a form of marketing that delivers advertising information through short-term (usually 15 seconds to 1 minute) video content. With the popularization of social media and mobile Internet, short video ads have become an important means of brand promotion due to their high interactivity and easy dissemination. Compared with traditional ads, short video ads have stronger visual impact and emotional resonance, and can quickly capture the audience's attention through creativity, humor, storytelling and other means. In addition, short video ads can be precisely targeted to help brands achieve more efficient marketing results among specific audiences.

[0003] At present, in order to increase the number of ad views, traditional short video ads are usually inserted into short video apps in the form of splash screen ads, information flow ads, and floating ads. Since the content of the ads does not match the current needs of users, they appear frequently when users open the app at any time, causing users to feel impatient when watching ads that are not related to their interests, which interferes with the user's viewing experience and reduces the user's favorability towards the products recommended in the ads. Summary of the invention

[0004] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. 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 creative work are within the scope of protection of the present invention.

[0005] The purpose of the present invention is to solve the shortcomings of the prior art and to propose an advertisement delivery system and delivery method based on data recognition.

[0006] In order to achieve the above-mentioned object, in a first aspect, the present invention provides an advertisement delivery system based on data identification, comprising:

[0007] The data collection module is used to collect user behavior data on the short video platform to form a behavior data set;

[0008] The data processing module is used to pre-process and filter the user behavior data in the data set through statistical algorithms, and match the push scenarios according to the advertising format;

[0009] The database module is used to store the advertisements to be delivered and to establish a push advertisement database according to the push advertisement format;

[0010] The advertisement delivery adjustment module is used to divide the correspondence between advertisement delivery time periods and advertisement delivery forms, and deliver advertisements according to push scenarios.

[0011] In some embodiments, the user behavior data includes user classification data and user behavior data, and the user behavior data is collected by a data collection module and transmitted to a data processing module.

[0012] In some embodiments, the user classification data includes age dimension, gender dimension, region dimension, preference dimension and consumption dimension, and the user behavior data includes participation behavior, emotional behavior and consumption behavior.

[0013] In some of the embodiments, the push advertisement forms include advertisements for daily necessities, clothing and accessories, digital appliances, and leisure and entertainment.

[0014] In some embodiments, the advertisement delivery time period includes a working time period and a leisure time period, the working time period includes a morning period and an afternoon period, and the leisure time period includes a midday period, an evening period, and a night period.

[0015] In some of the embodiments, the advertising delivery forms include opening screen display type advertisements, video integration type advertisements, dynamic reminder type advertisements and video playback type advertisements.

[0016] In a second aspect, the present invention further provides an advertisement delivery method based on data identification, which is executed via the advertisement delivery system based on data identification as described in the first aspect, and the delivery method steps include:

[0017] S100, collecting user behavior data on the short video platform through the short video platform to form a behavior data set;

[0018] S200, preprocessing and screening the user behavior data in the data set by using a statistical algorithm, and analyzing and obtaining user classification types and user behavior types;

[0019] S300, reading and storing the advertisements to be delivered, and establishing a push advertisement database according to the push advertisement format;

[0020] S400, reading the push advertisement form, and matching the push advertisement database advertisement data according to the user classification type and the user behavior type;

[0021] S500, dividing the advertising delivery time period and the advertising delivery form, and matching and pushing advertising data in the advertising database according to the user classification type and the user behavior type;

[0022] S600, reading user classification type and user behavior type data, and matching corresponding advertisement data in a push advertisement database.

[0023] In some embodiments, the step S200 includes:

[0024] S210, preprocessing the user behavior data in the data set by using a statistical algorithm to remove duplicate, invalid and abnormal data;

[0025] S220, using a machine learning algorithm to analyze and filter the user classification data and the user behavior data to obtain a user classification type and a user behavior type;

[0026] S230, establishing a user data set corresponding to the user classification type and the user behavior type, and waiting for matching push scenarios.

[0027] In some embodiments, the step S300 includes:

[0028] S310, reading and storing advertisements to be delivered, and classifying and storing different types of advertisements according to the form of push advertisements;

[0029] S320, establish a push advertisement database and wait for push advertisements.

[0030] In some embodiments, the step S500 includes:

[0031] S510, dividing the advertising delivery time period and establishing the advertising delivery push scenario;

[0032] S520, mapping different advertising time periods to corresponding advertising delivery forms, and waiting for advertisement delivery.

[0033] The present invention has the following beneficial effects:

[0034] The present invention, by dividing the whole day into different time periods, can adopt differentiated advertising delivery strategies according to user classification types and user behavior types; during the lunch and evening periods, users are mostly in a resting state, and the usage of short video APPs is relatively high, so splash screen ads and information flow ads can be selected for delivery. Information flow ads are similar to ordinary video content, and users can skip them by sliding, so they are more easily accepted during peak hours; at the same time, the viewing interference caused by the simultaneous appearance of splash screen ads and floating ads is avoided, thereby improving the user viewing experience; in other time periods, users are usually at work or study, and the frequency and duration of using short video APPs are relatively low. At this time, the use of splash screen ads and floating ads is more likely to attract the user's attention; by reasonably planning the delivery time and form of advertisements, not only can the exposure of advertisements be guaranteed, but also the interference with the user's viewing experience can be effectively reduced. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 This is a principle framework diagram of the advertising delivery system based on data identification proposed by the present invention;

[0036] Figure 2 The process of the advertising delivery method based on data identification proposed by the present invention Figure 1 ;

[0037] Figure 3 The process of the advertising delivery method based on data identification proposed by the present invention Figure 2 ;

[0038] Figure 4 The process of the advertising delivery method based on data identification proposed by the present invention Figure 3 ;

[0039] Figure 5 The process of the advertising delivery method based on data identification proposed by the present invention Figure 4 ;

[0040] Figure 6 This is a schematic diagram of the advertisement delivery method based on data identification proposed in the present invention.

[0041] Legend:

[0042] 1. Data collection module; 2. Data processing module; 3. Database module; 4. Advertising placement adjustment module. DETAILED DESCRIPTION

[0043] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. 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 creative work are within the scope of protection of the present invention.

[0044] The embodiment of the present application provides an advertising delivery system and delivery method based on data identification, which solves the problem that short video ads in the prior art are often automatically inserted when users watch videos, which makes the delivery time of ads more random and cannot accurately distinguish the user's usage scenarios and time periods; the excessive frequency and diversity of advertising content may make users feel bored, and even produce "advertising fatigue", which in turn affects the acceptance of ads and the viewing experience. However, by dividing the different time periods throughout the day, the present application can adopt differentiated advertising delivery strategies according to user classification types and user behavior types; by reasonably planning the delivery time and form of ads, it can not only ensure the exposure of ads, but also effectively reduce interference with the user's viewing experience.

[0045] Please refer to the following examples for details:

[0046] Reference Figure 1 and Figure 6 The present invention provides an embodiment of an advertisement delivery system based on data identification, the specific structure of which includes: a data acquisition module 1, a data processing module 2, a database module 3 and an advertisement delivery adjustment module 4, wherein:

[0047] (1) Data collection module 1 generates a comprehensive user behavior data set by collecting user behavior data on the short video platform, covering multiple dimensions of user behavior such as viewing time, interaction frequency, search history, comments and likes; this data set can reflect the activity and preferences of different user groups in different time periods, thereby providing data support for targeted advertising.

[0048] Exemplarily, the core task of the data collection module 1 is to obtain user behavior data from multiple channels, including short video platforms, social media, user devices, etc., to ensure that the collected data is multi-dimensional and comprehensive. Therefore, it is necessary to use the built-in tracking tools of the short video platform, third-party data tracking tools, and sensors of mobile devices to read data.

[0049] Data processing module 2 pre-processes and screens the collected raw data, removing invalid data, filling missing values, normalizing data, etc. It also screens out user behavior data that meets specific conditions based on the needs of the advertising format and delivery scenario, and accurately identifies the interests and usage habits of different user groups through behavioral pattern analysis, making precise matches for subsequent advertising push.

[0050] For example, the original data usually has problems such as noise, missing values, and duplicate data. These problems are solved in the data preprocessing stage, including: deduplication and outlier processing: exclude duplicate and abnormal data records to ensure data quality; missing value filling: for missing data, use mean filling, interpolation and other technologies to supplement; data normalization: standardize data of different dimensions to facilitate subsequent analysis. Then, based on large-scale user behavior data, regression algorithms, classification algorithms, clustering algorithms, deep learning and other technologies are applied to train models, mine users' potential interests, and build dynamic interest portraits of users through algorithm models, including interest areas, active time periods, interaction methods, viewing preferences and other information. These portraits can reflect the personalized needs of users and provide a basis for advertising decisions.

[0051] Database module 3 is responsible for storing the advertisement content to be delivered, and establishing a structured advertisement database according to different advertisement forms (such as splash screen advertisement, information flow advertisement, floating advertisement, etc.). The database not only contains basic advertisement information (such as advertisement type, material, delivery time, etc.), but also contains the matching rules between advertisement and user group characteristics, which provides a basis for advertisement delivery scheduling and ensures the timeliness and relevance of advertisement information.

[0052] The advertisement delivery adjustment module 4 is responsible for the specific execution of advertisement delivery, including dividing different advertisement delivery time periods, and formulating the correspondence between advertisement forms and specific time periods, and automatically matching the most appropriate advertisement forms and advertisement delivery time according to the analysis results of push scenarios and user behavior data; during peak hours (such as noon and evening), this module tends to choose advertisement forms that users are more receptive to, such as information flow advertisements, while during off-peak hours, it gives priority to splash screen advertisements or floating advertisements that can accurately capture user attention; through this intelligent adjustment mechanism, advertisement delivery can not only achieve the maximum exposure at the most appropriate time, but also ensure that the interference of advertisements on user viewing experience is minimized, thereby improving the overall effect of advertisement delivery.

[0053] Furthermore, the user behavior data includes user classification data and user behavior data, and the user behavior data is collected by the data collection module 1 and transmitted to the data processing module 2.

[0054] Specifically, user classification data includes:

[0055] (1) Age dimension: Children (0-12 years old) rely on adults to make consumption decisions and are mainly interested in entertainment (such as cartoons, simple games), education (enlightenment books, learning software) and food (children's snacks). Teenagers (13-19 years old) are very interested in fashion (trendy clothing, shoes), electronic entertainment (mobile games, e-sports equipment), social (social software, offline gathering places), etc. Youth (20-39 years old) have many demands for career development (training courses, industry information), quality of life (household goods, high-end catering), leisure and entertainment (travel, fitness), etc. Middle-aged people (40-59 years old) are more rational in their purchasing behavior and tend to buy large-ticket items such as insurance and real estate for their families. At the same time, they invest more in health and wellness (health products, fitness equipment), children's education (tuition, extracurricular tutoring) and other fields. Elderly people (60 years old and above) have consumption needs mainly concentrated in health care (drugs, medical devices), elderly care services (elderly communities, housekeeping services) and leisure and entertainment (elderly travel, interest groups).

[0056] (2) Gender dimension: Male users pay more attention to sports (sports events, sports equipment), technology (electronic products, smart hardware), automobiles (automotive information, automotive supplies) and other fields, and pay more attention to product functions, performance and technical parameters in the purchase decision process; female users focus on fashion and beauty (clothing, cosmetics), home life (home decoration, kitchenware), mother and baby (pregnancy products, baby products for women of childbearing age), and pay more attention to product appearance, quality, brand image and shopping experience;

[0057] (3) Geographical dimension: Urban users (including first-tier, second-tier, and third-tier cities) are exposed to a wealth of information and pay attention to emerging things (such as shared economy products and new consumption models), cultural entertainment (movies, concerts), and high-end consumption (luxury goods and high-end customized services); township and rural users pay more attention to daily necessities (agricultural supplies and basic daily necessities), local specialty products (agricultural products and handicrafts), and cost-effective products;

[0058] (4) Hobbies and interests, including cultural and artistic activities (such as visiting museums and watching plays), sports activities (such as running, basketball, and yoga) and fitness programs (gym exercise and outdoor fitness), electronic products (such as smart phones, computers, and smart wearable devices) and technology information (technology exhibitions and technology forums), local specialties (restaurant visits), cooking (learning cooking skills, buying ingredients and kitchenware), and food tasting (wine, coffee, etc.);

[0059] (5) Purchasing behavior dimension: Impulsive buyers are easily influenced by external factors (such as advertising, promotional activities, and recommendations from others) when shopping. They have a short purchasing decision-making process and may purchase frequently. Rational buyers conduct sufficient research before purchasing goods or services, compare the advantages and disadvantages of different brands and products, focus on product quality, cost-effectiveness, and after-sales service, and have a low purchasing frequency. High-spending buyers have strong financial strength, pursue high-quality, high-end products and services, are less sensitive to prices, and focus on brand image, personalized customization, and unique consumer experience. Price-sensitive buyers pay close attention to price factors during the purchasing process, look for products with the best cost-effectiveness, tend to buy discounted goods and affordable brands, and respond more positively to promotional activities.

[0060] User behavior data includes:

[0061] (1) Engagement-related behaviors: including likes, comments, shares, play duration, and pauses:

[0062] (2) Classification based on time dimension: including immediate behavior, continuous behavior, positive emotional behavior, negative emotional behavior, etc.

[0063] Correspondingly, push advertising forms include:

[0064] (1) Daily necessities ads: ads that are closely related to people’s daily lives, including ads for fresh food (such as vegetables, fruits, and meat), which mainly emphasize the freshness, origin advantages, and organic or green certification of the products; ads for processed food (such as canned food, snacks, and convenience foods) focus on selling points such as diverse flavors, brand awareness, nutritional value, or convenience; they cover a wide range, involving various basic necessities in daily life.

[0065] (2) Clothing and accessories advertising: targeting clothing, footwear, accessories and other products. They can be divided into fashion clothing advertising, which emphasizes the design style (such as retro style, trendy street style), brand concept, wearing occasions (business formal wear, casual sportswear), etc. of clothing; footwear advertising will highlight the novelty of the shoes, comfort (such as the use of advanced sole technology, soft insoles) and functionality (such as waterproof, breathable); accessories advertising (including jewelry, hats, scarves, etc.) focuses on displaying the design beauty, material (such as pure gold, gemstones), and matching effects of accessories to attract consumers to buy in order to enhance their overall image.

[0066] (3) Digital and electrical advertisements: For digital products, such as smartphone advertisements, they emphasize their powerful processor performance, high-definition screen display, excellent camera function and other technical parameters. They also showcase the software ecosystem (such as the richness of the app store) and the brand's unique features (such as facial recognition technology).

[0067] (4) Leisure and entertainment advertisements: attract consumers to participate in various leisure and entertainment activities. Including tourism advertisements, promoting the natural scenery of tourist destinations (such as beautiful beaches, magnificent mountains), cultural landscapes (such as historical sites, folk culture), and tourism services (such as luxury hotel accommodation, special food experience), etc.;

[0068] Furthermore, the advertising delivery time period includes:

[0069] (1) Working hours: The morning period (9:00-12:00) is when most people are working or studying. Users are more focused on their work tasks or learning content, and pay less attention to ads. Ads placed during this period should be concise and clear, and be able to quickly convey key information. The afternoon period (13:00-17:00) is similar to the morning period. People are still working or studying, but their attention may decrease in the afternoon. It is recommended to place ads that can help users refresh themselves or provide entertainment, as well as ads related to leisure activities after get off work.

[0070] (2) Leisure time: During the lunch hour (12:00-13:00), users have more time to browse their phones or other devices. You can place food and catering related ads, such as discount information for nearby restaurants and special food delivery ads. It is also suitable for casual game ads and popular life video recommendation ads on short video platforms. The evening hour (19:00-22:00) is a golden time for advertising, suitable for all types of ads. During the night hour (22:00-6:00 the next day), users' behavior is more differentiated during this period. Some people are ready to go to sleep, so it is suitable to place some sleep-aid product ads (such as sleep-aid aromatherapy and comfortable bedding). For night owls, you can place late-night food delivery ads, online reading platform ads, and night live broadcast platform ads.

[0071] Specifically, the forms of advertising include:

[0072] (1) Pre-roll display ads: Splash screen ads are full-screen ads when the application is launched. Its advantages are mandatory and high exposure. When users open the application, the splash screen ads will immediately catch their eyes and instantly grab their attention. However, splash screen ads also have certain disadvantages. Due to their mandatory nature, some users may feel disturbed. If the ad takes too long to load or cannot be skipped, it may cause users to be disgusted.

[0073] (2) Content-integrated advertising: Information flow advertising is interspersed in the natural content flow that users browse. It is similar in form to the surrounding content and can be well integrated into the user's browsing experience. It can be precisely delivered based on the user's interests, behavioral habits, etc., and the click-through rate and conversion rate of the advertisement can be increased by matching the content that the user may be interested in.

[0074] (3) Floating display ads: Floating ads are ads that float on the screen. They can be a small icon, a small window, or a dynamic element. They have high visibility and flexibility and can always remain in a certain position on the screen while the user is browsing content, attracting the user's attention.

[0075] For example, based on the characteristics of different time periods when users use short video APPs, the form of advertising delivery is selected in a targeted manner: when users are in the morning, afternoon and night time periods, since most users are at work or study during these time periods, the frequency and duration of using short video APPs are relatively limited. At this time, a combination of splash screen ads and floating ads are used for delivery. Splash screen ads can strongly attract the user's attention when the user starts the app, and floating ads can provide continuous reminders during the use of the app, thereby effectively capturing the user's attention; in the lunch and evening periods, these two periods are rest time, and the usage activity of short video APPs is at its peak. In order to reduce the interference of advertising on users' short video viewing experience, splash screen ads and information flow ads are selected for delivery. Splash screen ads can display key information at the moment of startup, while information flow ads are very similar in form to ordinary video content, and users can easily skip them by sliding during browsing. This relatively friendly, low-interference form of advertising is more easily accepted by the majority of users during peak periods.

[0076] By making detailed divisions of the time periods throughout the day and matching them with appropriate forms of advertising, we can ensure that the ads receive sufficient exposure while minimizing the adverse effects on users' short video viewing experience, thus achieving an effective balance between advertising effectiveness and user experience.

[0077] Reference Figure 2-Figure 5 The present invention also provides an embodiment of an advertisement delivery method based on data identification, which is executed by an advertisement delivery system based on data identification as in the above embodiment, and the delivery method steps include:

[0078] S100, collecting user behavior data on the short video platform through the short video platform to form a behavior data set;

[0079] On short video platforms, user behavior data includes but is not limited to clicks, viewing time, likes, comments, shares, and follows. These behavior data can be collected in real time through the platform's API interface or through log files. The collected raw data will be stored in a database to provide a basis for subsequent data processing and analysis.

[0080] S200, preprocessing and screening the user behavior data in the data set by using a statistical algorithm, and analyzing and obtaining user classification types and user behavior types;

[0081] Please continue reading Figure 3 In this embodiment, step S200 includes:

[0082] S210, preprocessing the user behavior data in the data set by using a statistical algorithm to remove duplicate, invalid and abnormal data;

[0083] S220, using a machine learning algorithm to analyze and filter the user classification data and the user behavior data to obtain a user classification type and a user behavior type;

[0084] S230, establishing a user data set corresponding to the user classification type and the user behavior type, and waiting for matching push scenarios;

[0085] In the preprocessing stage, the original data needs to be cleaned first to remove duplicate records, invalid data (such as errors or null data) and outliers (for example, a large number of repeated behaviors of users in a short period of time may be caused by robot operations). Common data cleaning methods include deduplication, filling missing values, and standardization. This step helps to ensure the accuracy and credibility of the data; after the data preprocessing is completed, machine learning algorithms (such as cluster analysis, classification algorithms, etc.) are used to conduct in-depth analysis of user behavior data to identify the characteristics of different user groups. Users can be classified into different groups according to their behavior patterns, such as "highly active users", "potential users" or "dormant users". At the same time, their behaviors can also be divided into types, such as "browsing behavior", "purchasing behavior", "social behavior", etc. Common machine learning algorithms include K-means clustering, decision trees, support vector machines, etc.; the user classification and behavior types obtained from the analysis are combined with the actual user data to form a comprehensive user data set. This data set not only includes the basic information of users, but also includes their behavioral characteristics, so that in the subsequent advertising push process, the appropriate advertising content and delivery time can be accurately matched.

[0086] S300, reading and storing the advertisements to be delivered, and establishing a push advertisement database according to the push advertisement format;

[0087] Please continue reading Figure 4 In this embodiment, step S300 includes:

[0088] S310, reading and storing advertisements to be delivered, and classifying and storing different types of advertisements according to the form of push advertisements;

[0089] S320, establishing a push advertisement database and waiting for push advertisements;

[0090] During the advertising process, advertisements usually have various forms. According to the form of advertisements and target audiences, advertisements need to be classified and stored according to certain rules. These advertisement data are stored in the push advertisement database. An efficient advertisement push database is established to manage advertisements in a unified manner to ensure that advertisements of different forms and contents can be quickly retrieved and retrieved. This database can adopt a distributed storage method and be equipped with necessary retrieval and indexing functions to ensure that advertisement data can be quickly matched and accurately delivered during the advertisement push process.

[0091] S400, reading the push advertisement form, and matching the push advertisement database advertisement data according to the user classification type and the user behavior type;

[0092] According to the user's classification type and behavior type, the corresponding advertising data is matched from the push advertising database. If the user belongs to the "purchase behavior" type, purchase-related ads are pushed; if the user belongs to the "social behavior" type, social-related ads are pushed. This matching process is based on an algorithm model and can be performed through a rule engine or a recommendation system based on machine learning. The recommendation system analyzes multiple dimensions such as user historical behavior, interests, and geographic location to accurately recommend advertising content to each user.

[0093] S500, dividing the advertising delivery time period and the advertising delivery form, and matching and pushing advertising data in the advertising database according to the user classification type and the user behavior type;

[0094] Please continue reading Figure 5 In this embodiment, step S500 includes:

[0095] S510, dividing the advertising delivery time period and establishing the advertising delivery push scenario;

[0096] S520, mapping different advertising time periods to corresponding advertising delivery forms, and waiting for advertisement delivery;

[0097] The advertising time period is divided according to the user's active time and the target advertising effect. Based on data analysis, the system will select the best time window for advertising push, maximize the exposure and click-through rate of the advertisement, and ensure that the advertising delivery form is consistent with the preferences and behaviors of the target user group, thereby improving the advertising effect.

[0098] S600, reading user classification type and user behavior type data, and matching corresponding advertisement data in a push advertisement database.

[0099] Finally, it should be noted that the above is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, it is still possible for those skilled in the art to modify the technical solutions described in the aforementioned embodiments or to make equivalent substitutions for some of the technical features therein. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. An advertisement delivery system based on data recognition, characterized in that: include: A data collection module (1) is used to collect user behavior data on the short video platform to form a behavior data set; A data processing module (2) is used to pre-process and filter the user behavior data in the data set through a statistical algorithm, and match the push scenario according to the advertisement form; A database module (3) is used to store advertisements to be placed and to establish a push advertisement database according to the push advertisement format; The advertisement delivery adjustment module (4) is used to divide the corresponding relationship between the advertisement delivery time period and the advertisement delivery form, and deliver the advertisement according to the push scenario matching.

2. The advertising delivery system based on data identification according to claim 1, characterized in that: The user behavior data includes user classification data and user behavior data, and the user behavior data is collected by the data collection module (1) and transmitted to the data processing module (2).

3. The advertising delivery system based on data identification according to claim 1, characterized in that: The user classification data includes age dimension, gender dimension, region dimension, preference dimension and consumption dimension, and the user behavior data includes participation behavior, emotional behavior and consumption behavior.

4. The advertising delivery system based on data identification according to claim 1, characterized in that: The push advertising forms include advertisements for daily necessities, clothing and accessories, digital electronics, and leisure and entertainment.

5. The advertising delivery system based on data identification according to claim 1, characterized in that: The advertisement delivery time period includes a working time period and a leisure time period. The working time period includes a morning period and an afternoon period, and the leisure time period includes a midday period, an evening period, and a night period.

6. The advertising delivery system based on data identification according to claim 1, characterized in that: The advertising delivery forms include splash screen display ads, video integration ads, dynamic reminder ads and video playback ads.

7. An advertisement delivery method based on data identification, characterized in that: The delivery method is executed via the advertisement delivery system based on data identification according to any one of claims 1 to 6, and the delivery method comprises the following steps: S100, collecting user behavior data on the short video platform through the short video platform to form a behavior data set; S200, preprocessing and screening the user behavior data in the data set by using a statistical algorithm, and analyzing and obtaining user classification types and user behavior types; S300, reading and storing the advertisements to be delivered, and establishing a push advertisement database according to the push advertisement format; S400, reading the push advertisement form, and matching the push advertisement database advertisement data according to the user classification type and the user behavior type; S500, dividing the advertising delivery time period and the advertising delivery form, and matching and pushing advertising data in the advertising database according to the user classification type and the user behavior type; S600, reading user classification type and user behavior type data, and matching corresponding advertisement data in a push advertisement database.

8. The method for placing advertisements based on data identification according to claim 7, characterized in that: The step S200 includes: S210, preprocessing the user behavior data in the data set by using a statistical algorithm to remove duplicate, invalid and abnormal data; S220, analyzing and filtering the user classification data and the user behavior data using a machine learning algorithm to obtain a user classification type and a user behavior type; S230, establishing a user data set corresponding to the user classification type and the user behavior type, and waiting for matching push scenarios.

9. The method for placing advertisements based on data identification according to claim 7, characterized in that: The step S300 includes: S310, reading and storing advertisements to be delivered, and classifying and storing different types of advertisements according to the form of push advertisements; S320, establish a push advertisement database and wait for push advertisements.

10. The method for placing advertisements based on data identification according to claim 7, characterized in that: The step S500 includes: S510, dividing the advertising delivery time period and establishing the advertising delivery push scenario; S520, mapping different advertising time periods to corresponding advertising delivery forms, and waiting for advertisement delivery.

Images