Directional advertisement putting system based on rule engine

Abstract

The invention discloses a directional advertisement putting system based on a rule engine, and belongs to the technical field of telecommunication networks. Through mutual cooperation and cooperation of an advertisement library management module, a user behavior construction module, an advertisement matching engine module, a flow reporting module and a real-time monitoring module, advertisement pre-putting information is processed based on a personalized label, and an advertisement is rapidly matched and selected when a terminal requests the advertisement. The problems of low efficiency and lack of flexibility caused by the fact that an existing advertisement delivery system is limited to a single terminal and a mechanical delivery mode are solved, consumed time and labor cost are reduced, advertisements are delivered more accurately according to labels, changes of consumer behaviors or changes of market trends can be responded in time, and the service life of the system is prolonged. And thus, the advertisement resource allocation accuracy and conversion rate are improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of telecommunication networks, and in particular relates to a directional advertisement delivery system based on a rule engine. Background Art

[0002] Existing advertising delivery systems are often limited to a single media or terminal, and usually use traditional methods to purchase advertising space. The traditional advertising space purchase model usually relies on fixed time periods or locations. This method is not only inefficient but also lacks flexibility. Advertisers often need to book advertising space a long time in advance, which makes it impossible for them to quickly adjust advertising strategies according to market changes. Due to the non-real-time allocation of advertising resources, advertisers often miss the best time to deliver, resulting in a low return on advertising investment. The advertising optimization process is also a long process. From collecting data to analyzing results, to formulating new strategies, each step requires a lot of time and manpower costs. This means that advertisers cannot monitor advertising effects in real time and make corresponding adjustments, and cannot respond to changes in consumer behavior or changes in market trends in a timely manner, which in turn affects the accuracy and conversion rate of advertising.

[0003] With the development of big data and artificial intelligence technology, modern advertising delivery systems are moving towards a more intelligent and personalized direction. New advertising technologies allow advertisers to accurately target users based on their interests, preferences, geographic location, historical behavior and other factors, and achieve personalized recommendations for thousands of people. However, many existing systems still do not fully utilize these advanced technologies, or there are technical bottlenecks in the implementation process, which prevents them from fully realizing their potential. Summary of the invention

[0004] The purpose of the present invention is to address the problem that the existing advertising delivery systems are often limited to a single media or terminal and adopt the method of purchasing advertising space for delivery, which is not only inefficient but also lacks flexibility. A targeted advertising delivery system based on a rule engine is proposed. The present invention realizes the processing of advertising pre-delivery information based on personalized tags and quickly matches and selects advertisements when the terminal requests advertisements through the mutual coordination and cooperation of an advertising library management module, a user behavior construction module, an advertising matching engine module, a traffic reporting module and a real-time monitoring module. The problem of low efficiency and lack of flexibility caused by the existing advertising delivery system being limited to a single terminal and a mechanical delivery method is solved, the time and labor costs are reduced, and advertisements can be delivered more accurately according to tags. It can also respond to changes in consumer behavior or changes in market trends in a timely manner, thereby improving the accuracy and conversion rate of advertising resource allocation.

[0005] In order to solve the above technical problems, the present invention provides a targeted advertising delivery system based on a rule engine, comprising:

[0006] Ad library management module, responsible for advertisers to configure advertising materials and support fast retrieval;

[0007] User behavior building module, which is used to collect and analyze user behavior data and then build personalized tags for users;

[0008] Advertisement matching engine module, used to process the generation and storage of advertisement pre-delivery information based on personalized tags, and to quickly match and select advertisements when a terminal requests an advertisement;

[0009] Traffic reporting module, used to collect feedback data after advertising is delivered and evaluate the effectiveness of advertising based on the feedback data;

[0010] Real-time monitoring module, used to monitor the effectiveness of advertising and the operation status of monitoring services.

[0011] Preferably, the advertiser responsible for configuring the advertising material includes the advertiser responsible for configuring the type of advertisement slot, uploading the advertising material, the targeted group of people for advertisement delivery, and the time period for advertisement delivery.

[0012] Preferably, the configuration of the advertisement slot type includes the advertiser selecting a suitable advertisement slot type through a user-friendly interface, wherein the advertiser selects a suitable advertisement slot type according to the goal of the advertising campaign and sets relevant parameters.

[0013] Preferably, the advertising material uploading includes supporting advertising materials in file formats uploaded by advertisers and performing basic format verification; the advertising delivery targeting population includes helping advertisers to combine attributes of multiple dimensions to define target user groups through a multi-level selector.

[0014] Preferably, the user behavior building module collects the user's online behavior data through an application program interface integrated into different media and terminals, analyzes the user's preference interests based on the online behavior data, and updates the user's online behavior data in real time.

[0015] Preferably, the analysis of user preferences and interests based on the online behavior data specifically includes: removing invalid data, deleting empty records, duplicate records and data that does not meet format requirements, converting data in different formats into a unified format, splitting the data into different categories, and finally adapting user portrait generation strategies according to categories and adding user tags.

[0016] Preferably, the advertisement matching engine module performs data processing including the following steps:

[0017] S1. Process the data within the custom time period at the defined time node to extract the targeting information and the corresponding advertising information and load them into the memory;

[0018] S2. When the terminal initiates an advertisement request, matching qualified advertisement pre-delivery information in the memory based on the online behavior data, the targeting information and the corresponding advertisement information;

[0019] S3. Based on the matched pre-delivery information of the advertisements, further screen out advertisements that match the current advertisement slot type; if there are multiple advertisements that meet the conditions, select one based on the advertisement delivery method set by the advertiser;

[0020] S4. Return the selected advertisement data to the terminal and display the corresponding advertisement content.

[0021] Preferably, in S3, further screening out advertisements that match the current ad slot type includes formulating a dynamic bidding strategy by combining user value, ad inventory and competitor bidding factors, and adjusting and optimizing the bidding strategy through a real-time feedback mechanism based on the advertising delivery effect of the traffic reporting module.

[0022] Preferably, the evaluation of the advertisement effect based on the feedback data in the traffic reporting module includes using the Kafka message queue middleware for data processing, specifically including:

[0023] Sa. Count the number of times each ad is exposed, as well as the exposure trend within a specific time period;

[0024] Sb. further counting the number of clicks on each advertisement, and calculating the click-through rate in combination with the number of exposures;

[0025] Sc, calculate the cost per thousand impressions and cost per click based on the number of exposures and clicks to evaluate the effectiveness of the advertisement;

[0026] Sd. Adjust according to the results of effect evaluation.

[0027] Preferably, the monitoring service operation status in the real-time monitoring module includes monitoring the operation status of each module in the entire system, including CPU utilization, memory usage, network delay. If a system status is found to be abnormal, SMS and email alarms are received to handle the abnormality in time; the results of the effect evaluation reflected by the monitoring feedback data are obtained.

[0028] Compared with the prior art, the present invention has the following beneficial effects:

[0029] 1. This solution uses the audience's personalized tags to process the generation and storage of pre-advertising information, as well as the quick matching and selection of ads when the terminal requests them. The traditional purchase of media advertising space is transformed into the purchase of target groups, which improves the accuracy of advertising delivery. Advertisers can specify groups of people to deliver ads, and by analyzing user information to generate user interests, regions and other tags, they can specify such groups for delivery. Not only can it span different media and terminals, but it can also optimize advertising delivery strategies through real-time data analysis, and improve advertising interactivity and conversion rates. Such a system will greatly improve the efficiency of advertising delivery, better serve the needs of advertisers, and provide users with a more relevant and attractive advertising experience.

[0030] 2. This solution can not only span different media and terminals, but also optimize advertising strategies through real-time data analysis, improve advertising interactivity and conversion rate. Such a system will greatly improve the efficiency of advertising, better serve the needs of advertisers, and provide users with a more relevant and attractive advertising experience.

[0031] 3. This solution sets up an intelligent bidding engine based on multi-factor optimization, which can dynamically adjust the bidding strategy according to the user's real-time behavior, contextual information and competitor bids. It shortens the optimization cycle, achieves rapid response, improves the efficiency of media resource sales and traffic utilization, improves user experience, and increases user conversion rate and satisfaction. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 It is a schematic diagram of the overall structure of the present invention. DETAILED DESCRIPTION

[0033] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below with reference to the accompanying drawings and embodiments.

[0034] Example 1: Figure 1 As shown in the figure, the advertising system aims to provide efficient and accurate advertising delivery services. Its core components include advertising library management, user behavior analysis, advertising matching engine, traffic reporting and real-time monitoring modules. The system adopts a microservice architecture design, and each module communicates through an API interface to ensure the high availability and scalability of the system. Specifically, it includes:

[0035] The advertising library management module is responsible for configuring advertising materials for advertisers and supporting fast retrieval; the user behavior construction module is used to collect and analyze user behavior data and then build personalized tags for users; the advertising matching engine module is used to process the generation and storage of pre-advertising information based on personalized tags, and to quickly match and select ads when the terminal requests ads; the traffic reporting module is used to collect feedback data after advertising delivery and evaluate the effectiveness of advertising based on the feedback data; the real-time monitoring module is used to monitor the effectiveness of advertising delivery and the operation status of the monitoring service.

[0036] The advertising system in this embodiment uses the User-Agent string to identify the type of terminal used by the user, such as a smartphone, tablet computer, personal computer, smart TV, etc., and obtains the user's screen size and resolution through JavaScript, and sends advertisements suitable for the current device. At the same time, it adopts responsive web design to ensure that the advertising content can automatically adapt to different screen sizes and achieve cross-media and cross-terminal delivery.

[0037] When there are differences in ad sizes, the width and height in the terminal display of the ad system can support up and down floating. For example, if a picture with a length of 20 and a width of 20 is currently required, and a floating of 5 is allowed, eligible ad materials within the range of 15-25 can be filtered out in the ad library.

[0038] Specifically, being responsible for advertisers configuring advertising materials includes being responsible for advertisers configuring ad slot types, uploading advertising materials, targeting audiences for advertising delivery, and advertising delivery time periods.

[0039] Ad slot type configuration includes the advertiser selecting a suitable ad slot type. Through a user-friendly interface, the advertiser selects a suitable ad slot type based on the goals of the advertising campaign and sets relevant parameters.

[0040] It should be noted that this feature allows advertisers to select the appropriate ad slot type, such as startup ads, pause ads, screen saver ads, pop-up ads, etc. Through a user-friendly interface, advertisers can select the appropriate ad slot type according to the goals of the advertising campaign and set relevant parameters such as size, position, etc.

[0041] Through such settings, we can enrich the types of advertisements and the scope of adaptation, improve the advertiser experience, and further increase user conversion rate and satisfaction.

[0042] Specifically, the advertising material uploading includes supporting the advertising materials in the file format uploaded by the advertiser and performing basic format verification; the advertising delivery targeting population includes helping the advertiser to combine attributes of multiple dimensions to define the target user group through a multi-level selector.

[0043] It should be noted that the ad creative upload function supports advertisers to upload various types of ad creatives, including pictures, videos, etc. It supports common file formats and performs basic format verification. It also provides a preview function to ensure that the ad creative meets the expected design.

[0044] Advertisement targeting is targeted targeting of specific groups of people, helping advertisers define target audiences, including but not limited to age, gender, geographic location, interests and hobbies. A multi-level selector is created, which can configure the target population region, interests, current ad placement location (such as display at the beginning of a specified video), etc., allowing advertisers to flexibly combine multiple dimensions to define their target user groups. Data from the user behavior analysis module can also be integrated to provide more accurate population labels.

[0045] The Apache Kafka stream processing framework is used to target targeted advertising groups, which enables real-time data processing and dynamic updates of user profiles to ensure the timeliness of advertising push. In addition, data from third-party data providers is introduced and seamlessly integrated into user profiles through data cleaning and standardization processes. The data cleaning and standardization modules ensure that data from different sources can be processed and analyzed consistently.

[0046] Advertisement delivery time period setting means that advertisers are allowed to specify the delivery time window of the advertisement, including the start date, end date, specific time period of each day, etc. A time selector is designed to support setting specific time periods, and also provide some preset time templates for advertisers to choose from.

[0047] It should be noted that before configuring the ad slot type, the advertiser qualification certification and ad placement review mechanism are required to ensure that the ad content complies with the requirements of laws and regulations. This is a condition that every ad must meet before it is placed.

[0048] This solution uses the personalized tags of the audience to process the generation and storage of pre-ad placement information. The traditional purchase of media advertising space is transformed into the purchase of target groups, which improves the accuracy of advertising. Advertisers can specify groups to place ads. Not only can it cross different media and terminals, but it can also improve the interactivity and conversion rate of ads. Such a system will greatly improve the efficiency of advertising, better serve the needs of advertisers, and provide users with a more relevant and attractive advertising experience.

[0049] Example 2: Figure 1 As shown, the targeted advertising delivery system based on the rule engine includes:

[0050] The advertising library management module is responsible for configuring advertising materials for advertisers and supporting fast retrieval; the user behavior construction module is used to collect and analyze user behavior data and then build personalized tags for users; the advertising matching engine module is used to process the generation and storage of pre-advertising information based on personalized tags, and to quickly match and select ads when the terminal requests ads; the traffic reporting module is used to collect feedback data after advertising delivery and evaluate the effectiveness of advertising based on the feedback data; the real-time monitoring module is used to monitor the effectiveness of advertising delivery and the operation status of the monitoring service.

[0051] Specifically, the user behavior building module collects the user's online behavior data through application program interfaces integrated into different media and terminals, analyzes the user's preference interests based on the online behavior data, and updates the user's online behavior data in real time.

[0052] Among them, analyzing user preferences and interests based on the online behavior data specifically includes: removing invalid data, deleting empty records, duplicate records and data that do not meet format requirements, converting data in different formats into a unified format, splitting the data into different categories, and finally adapting user portrait generation strategies according to categories and adding user tags.

[0053] It is understandable that data from third-party data providers can be introduced and seamlessly integrated into user portraits through data cleaning and standardization processes. Through the data cleaning and standardization module, ensure that data from different sources can be processed and analyzed consistently. Unifying data formats is to convert data in different formats into a unified format, such as date format, currency format, etc. Data classification is to classify data, such as shopping, viewing, and personal social categories to split data into different categories. Data processing is to adapt the user portrait generation strategy according to the category and add user tags.

[0054] It should be noted that users' online behavior data is collected through application programming interfaces (APIs) integrated into different media and terminals, including but not limited to users' geographic location, browsing history, search history, purchase history, etc.

[0055] When analyzing online behavior, it is necessary to analyze the user's geographic location information and identify the user's permanent residence, recent visit, etc., so as to carry out regional targeted advertising. For example, if the user is often active in the Beijing area, Beijing-related advertisements can be displayed preferentially on media platforms in the Beijing area; advertisements around the user's permanent residence can also be prioritized based on the region.

[0056] It is also necessary to analyze the user's viewing history, including the type of content watched, viewing time, viewing frequency, and other information. By analyzing this data, it is possible to identify the user's interest preferences, such as whether the user often watches sports programs, technology news, etc., so as to construct the user's interest tags. If the user often watches sports programs, you can add sports interest tags when constructing this user's tags.

[0057] In reality, users’ interests and behaviors may change over time. The user behavior analysis module will regularly update user behavior data in real time to ensure the timeliness of user portraits. Here, user portraits can be understood as user interest tags.

[0058] This solution processes the pre-ad placement information through personalized tags constructed by the user behavior analysis module, and combined with the ad slot type configuration in Example 1, it can not only cross different media and terminals, but also optimize the ad placement strategy through real-time data analysis to improve the interactivity and conversion rate of ads. Such a system will greatly improve the efficiency of ad placement, better serve the needs of advertisers, and provide users with a more relevant and attractive ad experience.

[0059] Example 3: Figure 1 As shown, the targeted advertising delivery system based on the rule engine includes:

[0060] The advertising library management module is responsible for configuring advertising materials for advertisers and supporting fast retrieval; the user behavior construction module is used to collect and analyze user behavior data and then build personalized tags for users; the advertising matching engine module is used to process the generation and storage of pre-advertising information based on personalized tags, and to quickly match and select ads when the terminal requests ads; the traffic reporting module is used to collect feedback data after advertising delivery and evaluate the effectiveness of advertising based on the feedback data; the real-time monitoring module is used to monitor the effectiveness of advertising delivery and the operation status of the monitoring service.

[0061] Specifically, the advertisement matching engine module performs data processing including the following steps:

[0062] S1. Process the data within the custom time period at the defined time node to extract the targeting information and the corresponding advertising information and load them into the memory.

[0063] S2. When the terminal initiates an advertisement request, based on the online behavior data, the targeting information and the corresponding advertisement information, matching advertisement pre-delivery information that meets the conditions is performed in the memory.

[0064] S3. Based on the matching pre-ad placement information, further screen out advertisements that match the current ad slot type; if there are multiple eligible advertisements, select one based on the advertisement placement method set by the advertiser.

[0065] S4. Return the selected advertisement data to the terminal and display the corresponding advertisement content.

[0066] Specifically, in S3, advertisements that match the current ad slot type are further screened out, including combining user value, ad inventory, and competitor bidding factors, formulating a dynamic bidding strategy, and adjusting and optimizing the bidding strategy through a real-time feedback mechanism based on the advertising delivery effect of the traffic reporting module.

[0067] It is understandable that at the end of each day, the advertising system starts processing the data for that day, extracting all the targeting information and the corresponding advertising information and loading it into the memory. While loading the new data, the data from the previous day is cleaned up to avoid excessive memory usage.

[0068] When the terminal initiates an advertisement request, the advertising system first obtains the current user's regional information, interest preferences, currently watched program information, and the requested ad slot type. Based on the obtained user information, the pre-delivery information of eligible advertisements is quickly retrieved from the memory. From the retrieved pre-delivery information, advertisements that match the current ad slot type are further screened. If there are multiple eligible advertisements, the optimal advertisement is selected for display based on the advertisement delivery method set by the advertiser.

[0069] The selected advertisement data is returned to the terminal for displaying the corresponding advertisement content.

[0070] Specifically, in S3, advertisements that match the current ad slot type are further screened out, including combining user value, ad inventory, and competitor bidding factors, formulating a dynamic bidding strategy, and adjusting and optimizing the bidding strategy through a real-time feedback mechanism based on the advertising delivery effect of the traffic reporting module.

[0071] It should be noted that the dynamic bidding strategy combines multiple factors such as user value, advertising inventory, and competitor bids to formulate a dynamic bidding strategy;

[0072] The first is user value, which refers to the potential value of a user to an advertiser, and is usually evaluated based on factors such as the user's purchasing power, interest, and historical behavior. A detailed user profile is built based on the user's historical behavior data (such as browsing history, purchase history, and search history).

[0073] The second is advertising inventory, which refers to the number of advertising display opportunities that the advertising platform can provide. Use the stream processing technology Kafka to monitor the changes in advertising inventory in real time. According to the real-time changes in advertising inventory, dynamically adjust the bidding strategy. For example, when inventory is tight, increase the bid to ensure that the ad can be displayed.

[0074] Finally, competitor bids: Competitor bids refer to the bids of other advertisers on the same ad space. You can dynamically adjust your bids based on your competitors' bids to stay competitive.

[0075] When matching ads, the ad matching algorithm is used to match ads:

[0076] (1) Initialization and parameter verification: Get the current date, day of the week, and hour. Get the current time period. Initialize user and ad space data.

[0077] (2) Condition matching: Generate matching indexes based on user tags, regions, content types, and other conditions, and take the intersection of these indexes.

[0078] (3) Advertisement screening: Traverse the advertisement list and filter out eligible advertisements based on conditions such as advertisement status, bidding time limit, banned list, matching index, etc.

[0079] (4) Ad slot condition screening: Calculate the material size limit for each ad slot. Further screen the ads, taking into account conditions such as daily schedule, time schedule, actual available time, ad slot type, and latitude and longitude restrictions.

[0080] (5) Priority sorting: Sort by the priority of the advertisements and select the advertisement with the highest priority.

[0081] By matching through the ad matching algorithm, you can further filter out ads that match the current ad slot type.

[0082] In the advertising system, the traffic reporting module evaluates the effect of advertising based on feedback data, including using the Kafka message queue middleware for data processing, including:

[0083] Sa. Count the number of times each ad is exposed. Every time an ad is shown to a user, an exposure event occurs. The traffic reporting module records this event and counts it in the exposure statistics. If the user clicks on the ad, the traffic reporting module also records this behavior as an important basis for subsequent analysis and exposure trends within a specific time period.

[0084] Sb. Further count the number of clicks on each advertisement, and calculate the click-through rate in combination with the number of exposures.

[0085] Sc, calculate the cost per thousand impressions and cost per click based on the number of exposures and clicks to evaluate the effectiveness of the advertisement.

[0086] Sd. Make adjustments based on the results of the effectiveness evaluation to decide whether to continue running the current advertisement, whether it is necessary to adjust the budget allocation, or replace the creative materials, etc.

[0087] Specifically, the monitoring service operation status in the real-time monitoring module includes monitoring the operation status of each module in the entire system, including CPU utilization, memory usage, and network delay. If a system status is found to be abnormal, SMS and email alarms are received to handle the abnormality in time; the results of the effect evaluation reflected by the monitoring feedback data.

[0088] Through the real-time feedback mechanism, advertisers can continuously optimize their bidding strategies based on the results of advertising. Collect the effect data after advertising, including exposure rate, click-through rate, conversion rate, cost per click, cost per display, etc. Prometheus is used to collect system indicator data, and Grafana is used to draw data dashboards. Among them, the advertising revenue can be calculated based on the conversion rate, allowing advertisers to quickly change their delivery strategies. Real-time adjustments are achieved.

[0089] This solution sets up an intelligent bidding engine based on multi-factor optimization, which can dynamically adjust the bidding strategy according to the user's real-time behavior, contextual information and competitor bids. It shortens the optimization cycle, achieves rapid response, improves the efficiency of media resource sales and traffic utilization, improves user experience, and increases user conversion rate and satisfaction.

Claims

1. A targeted advertising delivery system based on a rule engine, characterized in that: Includes an ad library management module, which is responsible for advertisers to configure ad materials and support fast retrieval; User behavior building module, which is used to collect and analyze user behavior data and then build personalized tags for users; Advertisement matching engine module, used to process the generation and storage of advertisement pre-delivery information based on personalized tags, and to quickly match and select advertisements when a terminal requests an advertisement; Traffic reporting module, used to collect feedback data after advertising is delivered and evaluate the effectiveness of advertising based on the feedback data; Real-time monitoring module, used to monitor the effectiveness of advertising and the operation status of monitoring services.

2. The rule engine-based targeted advertising delivery system according to claim 1, characterized in that: The advertiser responsible for configuring the advertising material includes the advertiser responsible for configuring the advertising position type, uploading the advertising material, the target group for advertising delivery, and the advertising delivery time period.

3. The rule engine-based targeted advertising delivery system according to claim 2, characterized in that: The configuration of the advertisement slot type includes the advertiser selecting a suitable advertisement slot type. Through a user-friendly interface, the advertiser selects a suitable advertisement slot type according to the goal of the advertising campaign and sets relevant parameters.

4. The rule engine-based targeted advertising delivery system according to claim 3, characterized in that: The advertisement material uploading includes supporting the advertisement materials in the file format uploaded by the advertiser and performing format verification; the advertisement delivery targeting population includes helping the advertiser to combine attributes of multiple dimensions to define the target user group through a multi-level selector.

5. The rule engine-based targeted advertising delivery system according to claim 1, characterized in that: The user behavior building module collects the user's online behavior data through application program interfaces integrated into different media and terminals, analyzes the user's preference interests based on the online behavior data, and updates the user's online behavior data in real time.

6. The rule engine-based targeted advertising delivery system according to claim 5, characterized in that: The analyzing of user preferences and interests based on the online behavior data specifically includes: removing invalid data, deleting empty records, duplicate records and data that do not meet format requirements, converting data in different formats into a unified format, splitting the data into different categories, and finally adapting user portrait generation strategies according to categories and adding user tags.

7. The rule engine-based targeted advertising delivery system according to claim 1, characterized in that: The data processing by the advertisement matching engine module includes the following steps: S1. Process the data within the custom time period at the defined time node to extract the targeting information and the corresponding advertising information and load them into the memory; S2. When the terminal initiates an advertisement request, matching qualified advertisement pre-delivery information in the memory based on the online behavior data, the targeting information and the corresponding advertisement information; S3. Based on the matched pre-delivery information of the advertisements, further screen out advertisements that match the current advertisement slot type; if there are multiple advertisements that meet the conditions, select one based on the advertisement delivery method set by the advertiser; S4. Return the selected advertisement data to the terminal and display the corresponding advertisement content.

8. The rule engine-based targeted advertising delivery system according to claim 7, characterized in that: In S3, further screening out advertisements that match the current ad slot type includes formulating a dynamic bidding strategy by combining user value, ad inventory, and competitor bidding factors, and adjusting and optimizing the bidding strategy through a real-time feedback mechanism based on the advertising delivery effect of the traffic reporting module.

9. The rule engine-based targeted advertising delivery system according to claim 1, characterized in that: The evaluation of the advertisement effect based on the feedback data in the traffic reporting module includes using the Kafka message queue middleware to process the data, specifically including: Sa. Count the number of times each ad is exposed, as well as the exposure trend within a specific time period; Sb. further counting the number of clicks on each advertisement, and calculating the click-through rate in combination with the number of exposures; Sc, calculate the cost per thousand impressions and cost per click based on the number of exposures and clicks to evaluate the effectiveness of the advertisement; Sd. Adjust according to the results of effect evaluation.

10. The rule engine-based targeted advertising delivery system according to claim 1, characterized in that: The monitoring service operation status in the real-time monitoring module includes monitoring the operation status of each module in the entire system, including CPU usage, memory usage, network delay, and if a system status is found to be abnormal, SMS and email alarms are connected to handle the abnormality in time; The results of the effectiveness evaluation reflected by the monitoring feedback data.

Images