Advertisement preview analysis method, system, processor and storage medium

Abstract

The present invention discloses a method, system, processor and storage medium for advertisement preview analysis, comprising: segmenting advertisement content according to the characteristics of pre-delivered advertisement content, generating advertisement feature labels, and generating user behavior feature labels through a user behavior analysis module; evaluating the matching degree between each user behavior feature label and each advertisement feature label through a preset matching algorithm; calling an advertisement effect prediction model to provide a preliminary advertisement preview strategy according to the matching degree evaluation result; collecting user feedback information in real time through a data collection unit when the user interacts with the advertisement preview; and continuously optimizing and updating the advertisement effect prediction model; optimizing the preliminary advertisement preview strategy based on the optimized and updated advertisement effect prediction model to generate corresponding advertisement delivery strategy suggestions, and the continuous optimization process enables the advertisement content to significantly improve user satisfaction and bring higher user responsiveness to the advertisement.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to a method, system, processor and storage medium for advertisement preview analysis. Background Art

[0002] With the continuous development of big data and artificial intelligence technology, the advertising industry is undergoing a profound transformation. Advertisers and marketers are looking for more advanced ways to improve the conversion rate and return on investment of advertisements. In addition, the importance of user experience is increasing, and the personalized demand for advertising content and the accurate prediction of preview effects have become an indispensable part of marketing strategy.

[0003] Existing ad preview technologies mainly focus on simple user grouping based on user historical behavior data, and then displaying relevant ad previews. Although this method has improved the targeting of ads to a certain extent, it still has obvious defects: first, user grouping is too static and lacks real-time response to the evolution of user interests; second, the generation and optimization of ad previews lack personalization and fail to fully consider the user's specific preview scenarios and device characteristics, resulting in fluctuations and uncertainty in ad effects. These shortcomings of existing technologies have resulted in the efficiency and effectiveness of advertising delivery being far from the ideal state.

[0004] In view of this, it is necessary to improve the advertisement preview technology in the prior art to solve the technical disadvantage of lack of personalized design in advertisement delivery. Summary of the invention

[0005] The purpose of the present invention is to provide a method, system, processor and storage medium for advertisement preview analysis to solve the above technical problems.

[0006] To achieve this object, the present invention adopts the following technical solutions:

[0007] A method for advertisement preview analysis, comprising:

[0008] According to the characteristics of the pre-delivered advertising content, the advertising content is segmented and advertising feature tags are generated;

[0009] By analyzing the user's historical browsing records, application usage habits and feedback data, a user behavior analysis module is activated, and a user behavior feature tag is generated through the user behavior analysis module;

[0010] Evaluate the matching degree between each user behavior feature tag and each advertisement feature tag through a preset matching algorithm;

[0011] Based on the matching evaluation results, the advertising effect prediction model is called to provide a preliminary advertising preview strategy;

[0012] When the user interacts with the ad preview, the user's feedback information is collected in real time through the data collection unit;

[0013] Utilizing the feedback data collected in real time to continuously optimize and update the advertising effect prediction model;

[0014] The preliminary advertisement preview strategy is optimized based on the optimized and updated advertisement effect prediction model to generate corresponding advertisement delivery strategy recommendations.

[0015] Optionally, the step of segmenting the advertisement content according to the characteristics of the pre-delivered advertisement content to generate advertisement feature tags may specifically include:

[0016] Classify the pre-launched advertising content according to its main types, and extract and analyze the key attributes of each type of advertising content;

[0017] Analyze the target group based on the key attributes of the advertisement to obtain the user profile that the advertisement hopes to reach;

[0018] Relate the collected key attributes of the advertisement and the target group information to the specific consumption context;

[0019] Generate preliminary advertising feature labels based on the classification, key attributes, user portraits and context-related information of the advertising content.

[0020] Optionally, the step of segmenting the advertisement content according to the characteristics of the pre-delivered advertisement content to generate advertisement feature tags may also include:

[0021] Verify the generated preliminary ad feature labels in a small-scale target user group and collect preliminary feedback information;

[0022] The preliminary advertisement feature tags are verified and optimized through the preliminary feedback information to generate advertisement feature tags, and the advertisement feature tags are integrated into a tag library that can be continuously updated.

[0023] Optionally, the method activates a user behavior analysis module by analyzing the user's historical browsing records, application usage habits and feedback data, and generates a user behavior feature tag through the user behavior analysis module; specifically includes:

[0024] Centrally collect the user's historical browsing records, application usage habits, and feedback data on historical advertising content to obtain the first data package;

[0025] Performing preliminary cleaning processing on the first data packet, wherein the cleaning processing includes removing invalid or erroneous data records;

[0026] Using deep machine learning data mining technology, we can identify the main behavior patterns of users from the first cleaned data packet;

[0027] In-depth analysis of user feedback data on advertising content to analyze user preferences and reaction intensity to different types of advertising content;

[0028] Activate the user behavior analysis module to generate a set of user behavior feature labels for each user based on the identified main behavior patterns and user feedback data;

[0029] According to the generated behavioral feature tags, users are grouped and their behavioral feature tags are regularly updated based on the users’ latest behavioral patterns and user feedback data using a real-time feedback loop mechanism.

[0030] Integrate all users' behavioral feature tags into a dynamically updated behavioral feature database.

[0031] Optionally, the matching degree between each of the user behavior feature labels and each of the advertisement feature labels is evaluated by a preset matching algorithm; specifically including:

[0032] Select a preset matching algorithm and set weights for user behavior feature tags and ad feature tags;

[0033] Using the selected matching algorithm, the matching degree between each user behavior feature label and each advertisement feature label is calculated;

[0034] Analyze the preliminary results of the matching evaluation operation and identify the highest-scoring user and advertisement pairings;

[0035] Adjust the weights of the user behavior feature labels and the advertisement feature labels to perform sensitivity analysis on the matching algorithm, evaluate the impact of label weight changes on the matching score, and optimize the matching algorithm;

[0036] After weight adjustment and matching algorithm optimization, re-evaluation calculation is performed to confirm the final matching score;

[0037] The final matching evaluation results are recorded and transmitted to the advertising effect prediction model.

[0038] Optionally, calling an advertisement effect prediction model to provide a preliminary advertisement preview strategy according to the matching evaluation result may specifically include:

[0039] Based on the matching evaluation result, the advertising effect prediction model is called, and a preliminary advertising preview strategy framework is constructed through the advertising effect prediction model;

[0040] Add user experience indicators to the preliminary ad preview strategy framework to generate a preliminary ad preview strategy;

[0041] An adjustment mechanism is set in the preliminary advertisement preview strategy, and an adjustment parameter is predefined. When user feedback data triggers the adjustment parameter, the adjustment mechanism is started to adjust the preliminary advertisement preview strategy.

[0042] Optionally, the advertising effect prediction model is continuously optimized and updated using the feedback data collected in real time; specifically, the following steps are included:

[0043] Tracking and collecting user feedback data on ad preview interactions through a data collection unit;

[0044] Summarize and organize the feedback data collected in real time, and use the feedback data as input variables for the optimized advertising effect prediction model;

[0045] Based on the feedback data collected in real time, evaluate the current performance indicators of the advertising effect prediction model, and formulate an optimization plan based on the evaluation results of the performance indicators;

[0046] The feedback data collected in real time is input into the advertising effect prediction model, and the advertising effect prediction model is retrained according to the optimization plan and the feedback data to continuously optimize and update the advertising effect prediction model.

[0047] The present invention provides an advertisement preview analysis system, which is used to implement the advertisement preview analysis method as described above; the advertisement preview analysis system specifically comprises:

[0048] An advertisement feature management unit, used to segment advertisement content and generate advertisement feature tags;

[0049] Data collection unit, used to collect user's historical browsing records, application usage habits and feedback data;

[0050] User behavior analysis module, used to generate user behavior feature tags based on user's historical browsing records, application usage habits and feedback data;

[0051] A data processing unit storing a preset matching algorithm and an advertising effect prediction model; the preset matching algorithm is used to evaluate the matching degree between each user behavior feature label and each advertising feature label, and the advertising effect prediction model is used to generate a preliminary advertising preview strategy according to the matching degree evaluation result;

[0052] A strategy optimization unit, used to continuously optimize and update the advertising effect prediction model according to the feedback data collected in real time;

[0053] The monitoring unit includes an interactive interface for monitoring and presenting the running status of the advertisement preview analysis system.

[0054] The present invention provides a processor, comprising a memory and at least one processor, wherein instructions are stored in the memory;

[0055] The processor calls the instructions in the memory, so that the processor executes the above-mentioned advertisement preview analysis method.

[0056] The present invention provides a storage medium, on which instructions are stored, and the instructions are used to implement the above-mentioned advertisement preview analysis method.

[0057] Compared with the prior art, the present invention has the following beneficial effects: when working, the pre-delivered advertisement content is segmented to generate detailed advertisement feature tags; by analyzing the user's historical browsing records, application usage habits and feedback data, the user behavior analysis module is activated to generate user behavior feature tags; a preset matching algorithm is used to carefully evaluate the matching degree between user feature tags and advertisement feature tags; based on the matching degree evaluation result, the advertisement effect prediction model is called to propose a preliminary advertisement preview strategy; the feedback data generated by the user's interaction with the preview is collected by the system in real time, which provides real-time data for the model, enabling it to optimize and update itself; finally, this The advertising effect prediction model after periodic optimization will adjust the preliminary advertising preview strategy and generate the final advertising delivery strategy recommendation; this method ensures a high degree of match between advertising and users through detailed user analysis and advertising content segmentation, thereby significantly improving the level of advertising personalization and user acceptance; the collection of real-time feedback and model optimization ensure that the dynamic adjustment of advertising strategies can quickly respond to changes in the market and user behavior, thereby improving the efficiency and effectiveness of advertising delivery; in addition, the continuous optimization process enables the advertising content to develop in sync with user behavior and preferences, which can significantly improve user satisfaction in the long run and bring higher user responsiveness to advertising. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0059] The structures, proportions, sizes, etc. illustrated in the drawings of this specification are only used to match the contents disclosed in the specification so as to facilitate understanding and reading by persons familiar with this technology. They are not used to limit the conditions under which the present invention can be implemented, and therefore have no substantive technical significance. Any structural modification, change in proportion or adjustment of size, without affecting the effects and purposes that can be achieved by the present invention, should still fall within the scope of the technical contents disclosed by the present invention.

[0060] Figure 1 This is a flowchart of the method for previewing and analyzing advertisements according to the first embodiment of the present invention;

[0061] Figure 2 This is a second flow chart of the method for advertisement preview analysis of the first embodiment;

[0062] Figure 3 This is a third flowchart of the advertisement preview analysis method of the first embodiment. DETAILED DESCRIPTION

[0063] In order to make the purpose, features and advantages of the present invention more obvious and easy to understand, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the embodiments described below 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.

[0064] In the description of the present invention, it should be understood that the terms "upper", "lower", "top", "bottom", "inside", "outside" and the like indicate positions or positional relationships based on the positions or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation of the present invention. It should be noted that when a component is considered to be "connected" to another component, it may be directly connected to the other component or there may be a centrally arranged component at the same time.

[0065] The technical solution of the present invention is further described below with reference to the accompanying drawings and through specific implementation methods.

[0066] Embodiment 1:

[0067] Combination Figures 1 to 3 As shown, an embodiment of the present invention provides a method for advertising preview analysis, including:

[0068] S1, based on the characteristics of the pre-delivered advertising content, the advertising content is segmented and advertising feature labels are generated; these advertising feature labels include not only the attributes of the product or service, but also the expected user response and emotional tendency;

[0069] This step focuses on understanding and segmenting ad content, taking into account not only the physical attributes of the product or service, but also the expected user response and emotional tendencies. This multi-dimensional labeling ensures that the ad preview system can accurately understand the uniqueness of the ad content and how it may resonate with a specific user group; this preliminary step is the foundation for building an effective advertising strategy.

[0070] S2, by analyzing the user's historical browsing records, application usage habits and feedback data, activates the user behavior analysis module, and generates user behavior feature tags through the user behavior analysis module; these user behavior feature tags reflect the user's interest preferences, active time periods and interaction preferences.

[0071] By analyzing users' ad browsing behavior, including browsing history, app usage habits and previous feedback, the system generates labels about user interests, activity cycles and interaction preferences; this allows the system to accurately draw the behavioral profile of each user, providing a basis for subsequent personalized ad previews.

[0072] S3, evaluating the matching degree between each user behavior feature label and each advertisement feature label through a preset matching algorithm;

[0073] In this step, a pre-set matching algorithm is used to evaluate the degree of match between user behavior feature tags and advertising feature tags; this evaluation aims to determine which advertising content is most likely to arouse the interest of specific user groups, thereby ensuring the relevance and efficiency of advertising delivery.

[0074] S4, based on the matching evaluation result, calling the advertising effect prediction model to provide a preliminary advertising preview strategy;

[0075] Based on the results of the matching evaluation, the advertising effect prediction model provides a preliminary advertising preview strategy; this strategy takes into account the expected effect of the advertisement and attempts to recommend the most appropriate advertising content, that is, the one that is most likely to generate a positive response, to the target users.

[0076] S5, when the user interacts with the ad preview, the user's feedback information is collected in real time through the data collection unit; the feedback information includes click-through rate, viewing time, and interactive behavior data;

[0077] When users interact with advertising content, the system collects data on click-through rate, viewing time and interactive behavior in real time; this real-time feedback information is an indispensable data source for optimizing advertising strategies and predictive models.

[0078] S6, using the feedback data collected in real time to continuously optimize and update the advertising effect prediction model;

[0079] Using the collected real-time feedback data, the system continuously optimizes and updates the advertising effect prediction model, which ensures that the model can learn and adapt to changes in user behavior and maintain the accuracy and efficiency of its predictive ability.

[0080] S7, optimizing the preliminary advertisement preview strategy based on the optimized and updated advertisement effect prediction model to generate corresponding advertisement delivery strategy recommendations.

[0081] Finally, based on the optimized and updated ad effect prediction model, the system further optimizes the initial ad preview strategy and generates more accurate ad delivery strategy recommendations; this ensures that the ad strategy continues to adapt to changes in user behavior over time and optimizes ad delivery results.

[0082] The working principle of the present invention is as follows: when working, the pre-delivered advertising content is segmented to generate detailed advertising feature tags, and the user behavior analysis module is activated to generate user behavior feature tags by analyzing the user's historical browsing records, application usage habits and feedback data; a preset matching algorithm is used to carefully evaluate the matching degree between user feature tags and advertising feature tags; based on the matching degree evaluation result, the advertising effect prediction model is called to propose a preliminary advertising preview strategy; the feedback data generated by the user's interaction with the preview is collected by the system in real time, which provides real-time data for the model, enabling it to perform self-optimization and update; finally, this periodically optimized advertising effect prediction model The model will adjust the preliminary ad preview strategy and generate the final ad delivery strategy recommendation; compared with the ad preview technology in the prior art, this method ensures a high degree of match between ads and users through detailed user analysis and ad content segmentation, thereby significantly improving the level of ad personalization and user acceptance; the collection of real-time feedback and model optimization ensure that the dynamic adjustment of ad strategies can quickly respond to changes in the market and user behavior, thereby improving the efficiency and effectiveness of ad delivery; in addition, the continuous optimization process enables ad content to develop in sync with user behavior and preferences, which can significantly improve user satisfaction in the long run, while bringing higher user responsiveness to ads.

[0083] In this embodiment, it is specifically described that, in combination with Figure 2 As shown, step S1 specifically includes:

[0084] S11, classifying the pre-delivered advertising content according to its main types, and extracting and analyzing key attributes of each type of advertising content;

[0085] Classify advertising content by its main types, such as video ads, image ads, and text ads; for each type of advertising content, further extract and analyze its key attributes, which include the visual elements, copy content, and call to action of the advertisement; this is a preliminary step in identifying and understanding the core characteristics of advertising content.

[0086] S12, based on the key attributes of the advertisement, analyze the target group and obtain the user profile that the advertisement is expected to reach;

[0087] After understanding the key attributes of the advertisement, this step aims to analyze the target user group (i.e., user portrait) based on the advertisement attributes. This includes analyzing the age, gender, interests, hobbies, and purchasing power characteristics of the users that the advertisement content is expected to reach. This helps to clarify the advertisement's positioning and target market.

[0088] S13, associate the collected key attributes of the advertisement and the target group information with specific consumption contexts;

[0089] Associate the collected key attributes of the advertisement and the target user group information with a specific consumption context; this step considers the specific environment and context in which the advertisement reaches the user, such as time, location, and device used. Contextual relevance is a key step in improving the appeal and effectiveness of advertisements because it helps ensure that the advertising content is highly relevant to the user's current context.

[0090] S14, generating preliminary advertising feature labels based on the classification, key attributes, user portraits and contextual association information of the advertising content.

[0091] Based on the previous analysis, a preliminary set of advertising feature labels is generated according to the classification, key attributes, target user portraits and contextual associations of the advertising content. These advertising feature labels will be used to describe the comprehensive characteristics of the advertisements, which are crucial for subsequent advertising matching analysis and advertising effect prediction.

[0092] S15, verifying the generated preliminary advertising feature labels in a small-scale target user group and collecting preliminary feedback information;

[0093] The generated preliminary ad feature labels are actually verified in a small-scale target user group. This process includes testing the attractiveness of the ads, user interaction behavior, and feedback collection. The feedback information in the verification phase is crucial to confirming the accuracy and effectiveness of the ad labels.

[0094] S16, verifying and optimizing the preliminary advertising feature labels through preliminary feedback information, generating advertising feature labels, and integrating the advertising feature labels into a continuously updateable label library.

[0095] Based on the preliminary feedback information collected in step S15, the initially generated ad feature tags are refined, verified and optimized. The optimized ad feature tags more accurately reflect the relevance between ad content and target users. Finally, these optimized ad feature tags are integrated into a continuously updated tag library for continuous use and reference by the ad preview analysis system.

[0096] In this embodiment, it is specifically described that, in combination with Figure 3 As shown, step S2 specifically includes:

[0097] S21, centrally collect the user's historical browsing records, application usage habits, and feedback data on historical advertising content to obtain a first data packet;

[0098] This step involves centrally collecting extensive data about users, including but not limited to historical browsing records, application usage habits, and user feedback on historical advertising content. These data constitute the first data package and are the basis for analyzing user behavior and optimizing advertising strategies.

[0099] S22, performing preliminary cleaning processing on the first data packet, the cleaning processing includes eliminating invalid or erroneous data records; eliminating occasional misclick operations, abnormal browsing behaviors; ensuring the data quality of subsequent analysis and processing, and improving the accuracy of the analysis.

[0100] The collected data often contains some noise, such as erroneous records, repeated information or irrelevant data. Cleaning the first data packet and removing invalid or erroneous records is a necessary step to ensure data quality and analysis accuracy.

[0101] S23, using deep machine learning data mining technology to identify the main behavior patterns of users from the first cleaned data packet;

[0102] This step uses deep machine learning and data mining technology to identify the user's main behavior patterns from the cleaned data, including the key behavioral characteristics of the user's browsing preferences, activity time, and frequency, providing high-quality input for generating user behavior labels.

[0103] S24, deeply analyzing the user feedback data on the advertising content to analyze the user's preference and reaction intensity to different types of advertising content;

[0104] By deeply analyzing user feedback on advertising content, we can abstract users’ preferences and reaction intensity to different types of advertising content, which is conducive to understanding users’ specific needs and feelings and providing a basis for personalized advertising recommendations.

[0105] S25, activating the user behavior analysis module, and generating a set of user behavior feature labels for each user based on the identified main behavior patterns and user feedback data;

[0106] Based on the identified user behavior patterns and user feedback data, the user behavior analysis module is activated to generate a set of detailed user behavior feature labels for each user.

[0107] S26, grouping the users according to the generated behavior feature tags, and using a real-time feedback loop mechanism to regularly update their behavior feature tags according to the users' latest behavior patterns and user feedback data;

[0108] Based on the generated behavioral feature tags, users are divided into different groups to achieve more detailed market segmentation and target positioning; at the same time, a real-time feedback loop mechanism is introduced to regularly update the behavioral feature tags based on the user's latest behavior patterns and feedback data. This dynamic update mechanism ensures the accuracy and timeliness of user tags.

[0109] S27, integrating all users' behavior feature tags into a dynamically updated behavior feature database.

[0110] All user behavior feature tags are integrated into a dynamically updated behavior feature database, which provides rich, real-time updated data resources for advertising prediction models and personalized recommendations.

[0111] In this embodiment, it is specifically described that step S3 specifically includes:

[0112] S31, selecting a preset matching algorithm and setting weights for user behavior feature tags and advertisement feature tags;

[0113] Select a suitable preset matching algorithm and set weights for user behavior feature tags and advertising feature tags. The weight setting is crucial because it determines which tag feature is more important in the matching process, thus affecting the final matching score. The weight setting reflects the relative importance of different features in the promotion strategy.

[0114] S32, using the selected matching algorithm, calculating the matching degree between each user behavior feature label and each advertisement feature label; wherein the matching degree calculation process is to set a matching degree scoring index to compare and obtain the matching degree (attractiveness and relevance) of each advertisement to the user;

[0115] The selected matching algorithm is used to calculate the matching degree between each user behavior feature label and the advertising feature label. By setting a matching score indicator, this process quantifies the attractiveness and relevance of each advertisement to different users, that is, the matching degree. This quantification process is the key to achieving personalized advertising recommendations.

[0116] S33, analyzing preliminary results of the matching evaluation operation, and identifying the highest-scoring user and advertisement pairing;

[0117] Analyze the preliminary results of the match evaluation operation to identify the highest-scoring user-ad pairs; these high-scoring pairs represent optimal advertising opportunities, indicating higher user engagement and advertising effectiveness.

[0118] S34, adjusting the weights of the user behavior feature labels and the advertisement feature labels to perform sensitivity analysis on the matching algorithm, to evaluate the impact of the label weight changes on the matching score, and to optimize the matching algorithm;

[0119] By adjusting the weights of user behavior feature labels and advertising feature labels, sensitivity analysis is performed to evaluate the impact of changes in label weights on the matching score. Through this step, it is possible to identify which weight adjustments can significantly improve the matching degree and optimize the matching algorithm accordingly, which helps to improve the sensitivity and accuracy of the matching algorithm.

[0120] S35, after the weight adjustment and the optimization of the matching algorithm, the matching evaluation operation is re-performed to confirm the final matching score;

[0121] After weight adjustment and matching algorithm optimization, the matching evaluation operation is performed again to confirm the final matching score; this step verifies the effect of the previous optimization measures, determines the optimized matching score, and ensures the effectiveness of the matching algorithm.

[0122] S36, recording the final matching evaluation result and transmitting it to the advertising effect prediction model.

[0123] The final matching evaluation results are recorded and transmitted to the advertising effect prediction model. This step ensures that the advertising effect prediction model runs based on the most accurate and optimized matching information, thereby improving the accuracy of the advertising prediction model and the efficiency of advertising delivery.

[0124] In this embodiment, it is specifically described that step S4 specifically includes:

[0125] S41, calling the advertising effect prediction model according to the matching evaluation result, and building a preliminary advertising preview strategy framework through the advertising effect prediction model;

[0126] The ad preview strategy framework allocates the ad content to highly matched user groups based on its expected effect, and preliminarily plans the ad display frequency, time period, and format.

[0127] S42, adding user experience indicators to the preliminary advertisement preview strategy framework to generate a preliminary advertisement preview strategy;

[0128] Ensure that ad preview strategies are based not only on match and predicted performance, but also on user preferences and acceptance.

[0129] Among them, the possible risk points in the preliminary advertising preview strategy are analyzed, such as users' potential aversion to certain advertising content, factors that may affect advertising effectiveness due to changes in the market environment, etc., to provide a basis for adjusting the strategy.

[0130] S43, setting an adjustment mechanism in the preliminary advertisement preview strategy, predefining an adjustment parameter, and when the user feedback data triggers the adjustment parameter, starting the adjustment mechanism to adjust the preliminary advertisement preview strategy.

[0131] Set up flexible adjustment mechanisms in the preliminary strategy to quickly adjust the ad preview strategy based on real-time data and feedback. This includes pre-defined adjustment parameters (such as response measures when the click-through rate drops to a certain level) to ensure the effectiveness and timeliness of the strategy.

[0132] Based on the predicted effect and user experience, the advertising resources are initially allocated. This includes allocating the advertising budget, selecting appropriate advertising channels and platforms, and planning the production and release cycle of advertising content.

[0133] In this embodiment, it is specifically described that step S6 specifically includes:

[0134] S61, tracking and collecting user feedback data on the interaction with the advertisement preview through a data collection unit;

[0135] Establish a data collection unit, which is the basis of the continuous optimization process. The data collection unit tracks and collects feedback data from users' interactions with ad previews. Feedback data includes key indicators such as click-through rate, viewing time, user comments and ratings. The establishment of a real-time monitoring system can not only quickly capture changes in user behavior, but also provide real-time data support for further analysis.

[0136] S62, summarizing and organizing the feedback data collected in real time, and using the feedback data as input variables of an optimized advertising effect prediction model;

[0137] Summarizing and organizing the feedback data collected in real time makes the huge amount of data orderly, easy to understand and analyze; the organized feedback data will serve as an important input variable for optimizing the advertising effect prediction model; this ensures that the advertising prediction model can be adjusted according to the latest market and user reactions, enhancing the model's adaptability and predictability.

[0138] S63, based on the feedback data collected in real time, evaluating the current performance indicators of the advertising effect prediction model, and formulating an optimization plan according to the evaluation results of the performance indicators;

[0139] Based on the collected real-time feedback data, the current performance indicators of the advertising effect prediction model are evaluated; this includes key performance indicators such as model accuracy and error rate. According to the evaluation results of these performance indicators, a targeted optimization plan is formulated. The optimization plan involves one or more measures such as adjusting the model structure, optimizing algorithm parameters, and introducing new feature variables, aiming to solve specific problems existing in the model.

[0140] S64, inputting the feedback data collected in real time into the advertising effect prediction model, and retraining the advertising effect prediction model according to the optimization plan and the feedback data, so as to continuously optimize and update the advertising effect prediction model.

[0141] Feedback data collected in real time is input into the advertising effect prediction model, and the model is retrained according to the previously formulated optimization plan; this step is the key to the continuous optimization process, which ensures that the model can learn and adapt based on the latest data, thereby continuously improving the accuracy and effectiveness of the prediction. The retrained model is rigorously tested and verified, and will only be used to replace or update the previous model after ensuring stable performance and significant improvement.

[0142] Embodiment 2:

[0143] The present invention also provides an advertisement preview analysis system, which is used to implement the advertisement preview analysis method of the first embodiment; the advertisement preview analysis system specifically includes:

[0144] An advertisement feature management unit, used to segment advertisement content and generate advertisement feature tags;

[0145] Data collection unit, used to collect user's historical browsing records, application usage habits and feedback data;

[0146] User behavior analysis module, used to generate user behavior feature tags based on user's historical browsing records, application usage habits and feedback data;

[0147] A data processing unit stores a preset matching algorithm and an advertising effect prediction model; the preset matching algorithm is used to evaluate the matching degree between each user behavior feature label and each advertising feature label, and the advertising effect prediction model is used to generate a preliminary advertising preview strategy according to the matching degree evaluation result;

[0148] Strategy optimization unit, used to continuously optimize and update the advertising effect prediction model based on the feedback data collected in real time;

[0149] The monitoring unit includes an interactive interface for monitoring and presenting the operating status of the advertisement preview analysis system.

[0150] Embodiment three:

[0151] The present invention also provides a processor, comprising a memory and at least one processor, wherein instructions are stored in the memory;

[0152] The processor calls the instructions in the memory so that the processor executes the advertisement preview analysis method of the first embodiment.

[0153] Embodiment 4:

[0154] The present invention also provides a storage medium, characterized in that instructions are stored on the storage medium, and the instructions are used to implement the advertisement preview analysis method as described in the first embodiment.

[0155] As described above, the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features thereof may be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for advertising preview analysis, characterized in that: include: According to the characteristics of the pre-delivered advertising content, the advertising content is segmented and advertising feature tags are generated; By analyzing the user's historical browsing records, application usage habits and feedback data, a user behavior analysis module is activated, and a user behavior feature tag is generated through the user behavior analysis module; Evaluate the matching degree between each user behavior feature tag and each advertisement feature tag through a preset matching algorithm; Based on the matching evaluation results, the advertising effect prediction model is called to provide a preliminary advertising preview strategy; When the user interacts with the ad preview, the user's feedback information is collected in real time through the data collection unit; Utilizing the feedback data collected in real time to continuously optimize and update the advertising effect prediction model; Optimize the preliminary ad preview strategy based on the optimized and updated ad effect prediction model to generate corresponding ad placement strategy recommendations; The step of segmenting the advertisement content according to the characteristics of the pre-delivered advertisement content and generating advertisement feature tags specifically includes: Classify the pre-launched advertising content according to its main types, and extract and analyze the key attributes of each type of advertising content; Analyze the target group based on the key attributes of the advertisement to obtain the user profile that the advertisement hopes to reach; Relate the collected key attributes of the advertisement and the target group information to the specific consumption context; Generate preliminary ad feature labels based on ad content classification, key attributes, user profiles, and contextual information; Verify the generated preliminary ad feature labels in a small-scale target user group and collect preliminary feedback information; Verifying and optimizing the preliminary advertising feature tags through the preliminary feedback information, generating advertising feature tags, and integrating the advertising feature tags into a continuously updateable tag library; The method of activating a user behavior analysis module by analyzing the user's historical browsing records, application usage habits and feedback data, and generating a user behavior feature tag through the user behavior analysis module specifically includes: Centrally collect the user's historical browsing records, application usage habits, and feedback data on historical advertising content to obtain the first data package; Performing preliminary cleaning processing on the first data packet, wherein the cleaning processing includes removing invalid or erroneous data records; Using deep machine learning data mining technology, we can identify the main behavior patterns of users from the first cleaned data packet; In-depth analysis of user feedback data on advertising content to analyze user preferences and reaction intensity to different types of advertising content; Activate the user behavior analysis module to generate a set of user behavior feature labels for each user based on the identified main behavior patterns and user feedback data; According to the generated behavioral feature tags, users are grouped and their behavioral feature tags are regularly updated based on the users’ latest behavioral patterns and user feedback data using a real-time feedback loop mechanism. Integrate all users' behavioral feature tags into a dynamically updated behavioral feature database; The method of evaluating the matching degree between each user behavior feature tag and each advertisement feature tag by using a preset matching algorithm specifically includes: Select a preset matching algorithm and set weights for user behavior feature tags and ad feature tags; Using the selected matching algorithm, the matching degree between each user behavior feature label and each advertisement feature label is calculated; Analyze the preliminary results of the matching evaluation operation and identify the highest-scoring user and advertisement pairings; Adjust the weights of the user behavior feature labels and the advertisement feature labels to perform sensitivity analysis on the matching algorithm, evaluate the impact of label weight changes on the matching score, and optimize the matching algorithm; After weight adjustment and matching algorithm optimization, re-evaluation calculation is performed to confirm the final matching score; The final matching evaluation result is recorded and transmitted to the advertising effect prediction model; The method of calling the advertisement effect prediction model to provide a preliminary advertisement preview strategy according to the matching evaluation result specifically includes: Based on the matching evaluation result, the advertising effect prediction model is called, and a preliminary advertising preview strategy framework is constructed through the advertising effect prediction model; Add user experience indicators to the preliminary ad preview strategy framework to generate a preliminary ad preview strategy; An adjustment mechanism is set in the preliminary advertisement preview strategy, and an adjustment parameter is predefined. When the user feedback data triggers the adjustment parameter, the adjustment mechanism is activated to adjust the preliminary advertisement preview strategy; The advertising effect prediction model is continuously optimized and updated using the feedback data collected in real time; specifically, the following steps are included: Tracking and collecting user feedback data on ad preview interactions through a data collection unit; Summarize and organize the feedback data collected in real time, and use the feedback data as input variables for the optimized advertising effect prediction model; Based on the feedback data collected in real time, evaluate the current performance indicators of the advertising effect prediction model, and formulate an optimization plan based on the evaluation results of the performance indicators; The feedback data collected in real time is input into the advertising effect prediction model, and the advertising effect prediction model is retrained according to the optimization plan and the feedback data to continuously optimize and update the advertising effect prediction model.

2. An advertisement preview analysis system, characterized in that: A method for implementing the advertisement preview analysis as claimed in claim 1; the advertisement preview analysis system specifically comprises: An advertisement feature management unit, used to segment advertisement content and generate advertisement feature tags; Data collection unit, used to collect user's historical browsing records, application usage habits and feedback data; User behavior analysis module, used to generate user behavior feature tags based on user's historical browsing records, application usage habits and feedback data; A data processing unit storing a preset matching algorithm and an advertising effect prediction model; the preset matching algorithm is used to evaluate the matching degree between each user behavior feature label and each advertising feature label, and the advertising effect prediction model is used to generate a preliminary advertising preview strategy according to the matching degree evaluation result; A strategy optimization unit, used to continuously optimize and update the advertising effect prediction model according to the feedback data collected in real time; The monitoring unit includes an interactive interface for monitoring and presenting the running status of the advertisement preview analysis system.

3. A processor, characterized in that: The device comprises a memory and at least one processor, wherein instructions are stored in the memory; The processor calls the instructions in the memory to enable the processor to execute the advertisement preview analysis method according to claim 1.

4. A storage medium, characterized in that: The storage medium stores instructions, and the instructions are used to implement the advertisement preview analysis method according to claim 1.

Images