Advertisement putting method, system, medium and program product based on big data analysis

Abstract

An advertisement putting method, system, medium and program product based on big data analysis, in the method, the acceptance degree data of target users to various advertisement contents is calculated; the consumption conversion link of target users in different time periods is determined based on purchase behavior data; the key advertisement content combination and the optimal display timing are determined according to the consumption conversion link; the advertisement contents with greater than a preset first similarity threshold and less than a preset frequency in the preset advertisement library are screened out, and an advertisement content exploration pool is obtained; the advertisement contents in the advertisement content exploration pool are inserted into the optimal display timing in a progressive manner to form a dynamic advertisement putting sequence; and the final advertisement contents corresponding to the dynamic advertisement putting sequence are put to the target users according to the dynamic advertisement putting sequence. The present application brings the advertisement contents meeting the potential interests of users, and realizes the continuous enhancement of the putting effect on the basis of maintaining the existing putting effect.

Description

TECHNICAL FIELD

[0001] The present application belongs to the field of digital data processing, and particularly relates to an advertisement delivery method and system based on big data analysis, a medium and a program product. BACKGROUND

[0002] With the continuous development of Internet technology and the rapid growth of user data, the accuracy and effectiveness of advertisement delivery are particularly important. Traditional advertisement delivery methods mainly rely on manual experience and simple user portraits for delivery decisions. This approach is not only inefficient, but also difficult to accurately grasp the actual needs of users, resulting in waste of advertising resources and unsatisfactory delivery results.

[0003] In the prior art, user interest models can be established by collecting user browsing records, search history, and shopping behavior data, and advertisement delivery can be performed based on model prediction results. This method can improve the accuracy of advertisement delivery to some extent, making the advertisement content more relevant to the actual needs of users.

[0004] However, excessive pursuit of surface data indicators such as user click rates when making advertisement delivery decisions can lead the system to preferentially push familiar or easily accepted advertisement content to users, limiting their exposure to new things and making advertisement delivery repetitive at a low level, which is difficult to stimulate users' potential interest in consumption, and thus affects the long-term effectiveness of advertisement delivery. SUMMARY

[0005] The present application provides an advertisement delivery method and system based on big data analysis, which is used to bring users advertisement content that meets their potential interests, and realizes continuous enhancement of delivery effectiveness on the basis of maintaining existing delivery effectiveness.

[0006] In a first aspect, the present application provides an advertisement delivery method based on big data analysis, which obtains historical click data and purchase behavior data of a target user. The historical click data includes clicked advertisement content, click time, and click frequency, and the purchase behavior data includes purchased product information, purchase time, and purchase amount.

[0007] The repeat click rate and average click interval of each type of advertisement content in the historical click data are calculated to obtain the acceptance degree data of the target user for each type of advertisement content.

[0008] Based on the purchase behavior data, the consumption conversion link of the target user in different time periods is determined. The consumption conversion link includes a sequence of advertisement content from the first click to the final purchase.

[0009] The key advertisement content combination and the optimal display timing are determined based on the consumption conversion link.

[0010] screening, from the preset advertisement library, advertisement contents with a combination similarity greater than a preset first similarity threshold and a click frequency less than a preset frequency, to obtain an advertisement content exploration pool;

[0011] inserting the advertisement contents in the advertisement content exploration pool into the optimal display timing in a progressive manner to form a dynamic advertisement delivery sequence;

[0012] delivering the final advertisement contents corresponding to the dynamic advertisement delivery sequence to target users according to the dynamic advertisement delivery sequence.

[0013] By adopting the above technical solutions, the historical click data and purchase behavior data of target users are obtained, the acceptance degree of users to advertisement contents is calculated in combination with the repeated click rate and the average click interval, the advertisement preference characteristics of users can be grasped, the key advertisement content combination and the optimal display timing are determined based on the consumption conversion link of users, the advertisement content sequence that can most effectively trigger the purchase behavior of users can be identified, the advertisement contents with high similarity but low display frequency are screened from the preset advertisement library to form an exploration pool, and the advertisement contents are inserted into the optimal display timing in a progressive manner, which not only maintains the effectiveness of advertisement delivery but also expands the range of advertisement experience of users, so that the advertisement delivery conforms to the acceptance habit of users, the freshness of contents is maintained, and the continuous enhancement of delivery effect is realized on the basis of maintaining the existing delivery effect.

[0014] In combination with some embodiments of the first aspect, in some embodiments, the key advertisement content combination and the optimal display timing are determined according to the consumption conversion link, specifically comprising:

[0015] statistically obtaining time distribution of each type of advertisement content triggering the click behavior of target users in the consumption conversion link to obtain time sensitivity data;

[0016] based on the time sensitivity data, calculating the conversion probability of different advertisement content combinations within a preset time window, the conversion probability representing the probability of the advertisement content combination triggering the purchase behavior of target users;

[0017] selecting the advertisement content combination with a conversion probability greater than a preset probability threshold as the key advertisement content combination;

[0018] analyzing the optimal display time interval of each advertisement content in the key advertisement content combination, establishing a display rule based on the display rule, and generating the optimal display timing based on the display rule.

[0019] By adopting the technical solution, the time distribution characteristics of the advertisement content triggering the user click behavior in the consumption conversion link are analyzed, the time sensitivity data of the user to different advertisement contents is obtained, and then the conversion probability of the advertisement content combination in a specific time window is calculated. The advertisement content combination with high conversion probability is selected as the key combination, and the optimal display time interval of each advertisement content in the key combination is analyzed to establish a display rule and generate an optimal display timing. The time preference characteristics of the user and the display rule of the advertisement content are combined to ensure that the advertisement content is displayed to the user at the most suitable time point, thereby improving the timeliness and pertinence of the advertisement delivery, and effectively improving the advertisement conversion rate.

[0020] In combination with some embodiments of the first aspect, in some embodiments, the advertisement contents in the advertisement content exploration pool are inserted into the optimal display timing in a progressive manner, specifically including:

[0021] The content similarity of each advertisement content in the advertisement content exploration pool and each advertisement content in the key advertisement content combination is calculated.

[0022] The advertisement contents in the advertisement content exploration pool are grouped according to the content similarity to obtain similar advertisement groups, and each similar advertisement group corresponds to an advertisement content in the key advertisement content combination.

[0023] In each similar advertisement group, the advertisement contents are sequentially inserted into the display timing point of the corresponding key advertisement content in the order of high to low content similarity.

[0024] By adopting the technical solution, the content similarity of each advertisement content in the advertisement content exploration pool and the key advertisement content combination is calculated, the advertisement contents in the exploration pool are grouped, and the advertisement contents are sequentially inserted into the display timing point of the corresponding key advertisement after the advertisement content in each similar advertisement group is inserted into the display timing point of the corresponding key advertisement content in the order of high to low content similarity. This progressive advertisement content insertion method forms a continuous transition in content between the newly added advertisement content and the original key advertisement content, maintains the overall relevance of the advertisement sequence, and smoothly introduces the new advertisement content into the user's attention range. Through the combination of similarity control and progressive insertion, the continuity and user experience of the advertisement delivery are improved, so that the advertisement sequence can maintain stable conversion effect and has content richness.

[0025] In combination with some embodiments of the first aspect, in some embodiments, after the final advertisement content corresponding to the dynamic advertisement delivery sequence is delivered to the target user according to the dynamic advertisement delivery sequence, the method further includes:

[0026] Real-time collection of feedback data of the target user, the feedback data including click behavior, dwell time and interactive operation;

[0027] Calculation of the user engagement score of the final advertisement content based on the feedback data;

[0028] remove the advertisement content whose user engagement score is lower than the preset engagement threshold from the advertisement content exploration pool.

[0029] By adopting the technical solutions described above, the feedback data of the user to the launched advertisement is collected in real time, including the click behavior, the stay duration and the interactive operation, the user engagement score of the advertisement content is calculated, and the advertisement content whose user engagement score is lower than the preset engagement threshold is removed from the exploration pool, so that the advertisement launching system can dynamically adjust the quality level of the advertisement content library and continuously optimize the advertisement content exploration pool. By timely cleaning the advertisement content with low user engagement, it is ensured that only high-quality advertisement content that can effectively attract the attention and interaction of the user is retained in the exploration pool, thereby improving the overall quality and effect of the advertisement launching.

[0030] In combination with some embodiments of the first aspect, in some embodiments, the user engagement score of the final advertisement content is calculated based on the feedback data, specifically including:

[0031] The click behavior, the stay duration and the interactive operation are respectively assigned with preset weight coefficients;

[0032] The comprehensive engagement index is calculated according to the feedback data, and the comprehensive engagement index is a weighted sum of the click behavior, the stay duration and the interactive operation and the corresponding weight coefficients;

[0033] The time decay coefficient of each final advertisement content is calculated based on the comprehensive engagement index;

[0034] The user engagement score is obtained by multiplying the comprehensive engagement index and the time decay coefficient.

[0035] By adopting the technical solutions described above, different weight coefficients are assigned to the click behavior, the stay duration and the interactive operation, so that the actual participation degree of the user to the advertisement content can be more comprehensively evaluated, and the evaluation deviation caused by a single index can be reduced. When calculating the comprehensive engagement index, the weighted sum is adopted, which can reflect the differentiated contribution of different interactive behaviors to the user engagement. The introduction of the time decay coefficient takes into account the feature that the user interest changes over time, so that the recent user participation behavior has higher reference value. The user engagement score obtained by multiplying the comprehensive engagement index and the time decay coefficient reflects the multi-dimensional features of the user participation behavior and also embodies the influence of the time factor, so that the actual launching effect of the advertisement content can be more accurately quantified.

[0036] In combination with some embodiments of the first aspect, in some embodiments, after the advertisement content whose user engagement score is lower than the preset engagement threshold is removed from the advertisement content exploration pool, the method further includes:

[0037] The remaining advertisement contents in the advertisement content exploration pool are subjected to cluster analysis to obtain subsets of advertisement contents of different theme categories;

[0038] calculate the quantity proportion of each subset of advertising content, and compare the quantity proportion with a preset balanced distribution proportion, the preset balanced distribution proportion including a target proportion corresponding to each subset of advertising content;

[0039] supplement advertising content with a quantity proportion less than a corresponding target proportion from a preset advertising library.

[0040] By using the above technical solutions, the clustering analysis of the remaining advertising content in the advertising content exploration pool can discover the distribution of advertising content of different theme categories. By calculating the quantity proportion of each subset of advertising content and comparing the quantity proportion with a preset balanced distribution proportion, the unbalanced distribution of advertising content can be identified. According to the comparison result of the actual proportion and the target proportion of each category of advertising content, the advertising content with insufficient quantity can be supplemented from the preset advertising library, so that the reasonable distribution of different theme categories in the advertising content exploration pool can be maintained, the diversity of advertising content is maintained, the over-concentration or serious shortage of advertising content of certain theme categories is avoided, the advertisement delivery can cover a wider range of user interest points, and the overall advertisement delivery effect is improved.

[0041] In some embodiments of the first aspect, the advertising content with a quantity proportion less than a corresponding target proportion is supplemented from a preset advertising library, and specifically includes:

[0042] determining the advertising content with a quantity proportion less than a corresponding target proportion as the advertising content to be supplemented;

[0043] calculating the difference between the quantity proportion of the advertising content to be supplemented and the corresponding target proportion;

[0044] matching the candidate advertising content corresponding to the advertising content to be supplemented in the preset advertising library;

[0045] calculating the similarity between the candidate advertising content and the existing content in the corresponding subset of advertising content;

[0046] selecting the candidate advertising content with a similarity less than a preset second similarity threshold as the supplement content, and adding the supplement content to the advertising content exploration pool, the quantity of the supplement content being the same as the difference.

[0047] By adopting the technical solution, the specific difference between the quantity proportion of the to-be-supplemented advertisement content and the target proportion can be calculated to accurately determine the quantity of the advertisement content that needs to be supplemented. When matching the candidate advertisement content, the similarity between the candidate advertisement content and the existing content is calculated, and the candidate advertisement content with a similarity less than a preset second similarity threshold is selected as the supplemented content, so that the repeated content can be reduced when supplementing the advertisement content, and the difference between the supplemented content and the existing content is ensured. Moreover, by strictly controlling the supplement quantity, the quantity of the advertisement of each type of theme in the advertisement content exploration pool can accurately reach the expected balanced distribution, and the quality of the advertisement content exploration pool is improved.

[0048] In a second aspect, the embodiments of the present application provide an advertisement delivery system based on big data analysis, which comprises one or more processors and a memory. The memory is coupled to the one or more processors, and is configured to store computer program codes. The computer program codes comprise computer instructions. The one or more processors invoke the computer instructions to enable the system to perform the method described in the first aspect and any possible implementation manner of the first aspect.

[0049] In a third aspect, the embodiments of the present application provide a computer readable storage medium comprising instructions, which, when executed on a system, cause the system to perform the method described in the first aspect and any possible implementation manner of the first aspect.

[0050] In a fourth aspect, the embodiments of the present application provide a computer program product, which, when executed on a system, causes the system to perform the method described in any possible implementation manner of the first aspect.

[0051] The one or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages:

[0052] 1. The present application provides an advertisement delivery method based on big data analysis. The historical click data and purchase behavior data of target users are obtained, the acceptance degree of the users to the advertisement content is calculated in combination with the repeated click rate and the average click interval, and the advertisement preference features of the users can be grasped. The key advertisement content combination and the optimal display timing are determined based on the consumption conversion link of the users, and the advertisement content sequence that can most effectively trigger the purchase behavior of the users can be identified. The advertisement content with high similarity but low display frequency is filtered from a preset advertisement library to form an exploration pool, and the exploration pool is inserted into the optimal display timing in a progressive manner. The effectiveness of the advertisement delivery is maintained, the range of the advertisement experience of the users is expanded, the advertisement delivery conforms to the acceptance habit of the users, the freshness of the content is maintained, the existing delivery effect is maintained, and the continuous enhancement of the delivery effect is realized.

[0053] 2. The application provides an advertisement delivery method based on big data analysis, which collects user feedback data on delivered advertisements in real time, including click behavior, dwell time and interactive operation, calculates user engagement scores of advertisement content, and removes advertisement content below a preset engagement threshold from the exploration pool, so that the advertisement delivery system can dynamically adjust the quality level of the advertisement content library and continuously optimize the advertisement content exploration pool. By timely cleaning up advertisement content with low user engagement, it ensures that only high-quality advertisement content that can effectively attract user attention and interaction is retained in the exploration pool, improving the overall quality and effectiveness of advertisement delivery.

[0054] 3. The application provides an advertisement delivery method based on big data analysis, which performs cluster analysis on the remaining advertisement content in the advertisement content exploration pool, and can discover the distribution of advertisement content of different theme categories. By calculating the number proportion of each subset of advertisement content and comparing it with the preset balanced distribution proportion, it can identify the unbalanced distribution of advertisement content. According to the comparison result of the actual proportion and the target proportion of each category of advertisement content, the number of insufficient advertisement content is supplemented from the preset advertisement library, which can maintain the reasonable distribution of different theme categories in the advertisement content exploration pool, maintain the diversity of advertisement content, avoid the over-concentration or serious shortage of advertisement content of certain theme categories, and make the advertisement delivery cover a wider range of user interest points, improving the overall advertisement delivery effect. BRIEF DESCRIPTION OF DRAWINGS

[0055] Figure 1 is a flowchart of an advertisement delivery method based on big data analysis in an embodiment of the application.

[0056] Figure 2 is a flowchart of an advertisement content optimization method based on feedback data in an embodiment of the application.

[0057] Figure 3 is an entity device structure diagram of an advertisement delivery system based on big data analysis provided in an embodiment of the application. DETAILED DESCRIPTION

[0058] The terms used in the following embodiments of the application are only for the purpose of describing the specific embodiments and are not intended to be limiting on the application. As used in the specification and the appended claims of the application, the singular forms "a," "an," and "the" are intended to include plural forms as well, unless the context clearly indicates otherwise. It should also be understood that the term "and / or" used in the application means any or all possible combinations of one or more listed items.

[0059] The terms "first", "second", "third", etc. are used only for descriptive purposes and do not connote or imply relative importance or a specific order. Thus, features defined with "first", "second" or "third" can include one or more of the features, and the description of the embodiments of the present application, unless otherwise stated, the meaning of "a plurality" is two or more.

[0060] The following describes an embodiment of the present application with one embodiment and in conjunction with Figure 1 An embodiment of the present application is described as follows:

[0061] Please refer to Figure 1 An embodiment of the present application is described as follows:

[0062] S101, obtaining the historical click data and purchase behavior data of the target user;

[0063] The system obtains the historical click data and purchase behavior data of the target user, the historical click data including the clicked advertisement content, the click time and the click frequency, and the purchase behavior data including the purchased product information, the purchase time and the purchase amount.

[0064] The system needs to obtain the historical click data and purchase behavior data of the target user as the input for subsequent analysis. Among them, the historical click data includes the advertisement content clicked by the user, the time and frequency of clicking, etc. These data reflect the user's interest and acceptance of different types of advertisements. The purchase behavior data includes the product information actually purchased by the user, the purchase time and the amount, etc. These data reflect the final effect of advertisement delivery, that is, whether the user is triggered by the advertisement and produces a purchase behavior.

[0065] The system can obtain these data through various channels, such as cooperating with advertisement platforms, e-commerce platforms, etc., accessing their user behavior log systems; or embedding data collection modules in self-owned products to record the click and purchase behavior of users in the products. For data from different sources, the system also needs to clean and integrate the data to ensure the accuracy and consistency of the data, so as to facilitate the comprehensive use of various data in subsequent analysis.

[0066] It should be noted that the collection of the historical click data and purchase behavior data of the target user in the present application is with the consent of the person and for legal scenarios.

[0067] S102, calculating the repeated click rate and average click interval of each type of advertisement content in the historical click data to obtain the acceptance degree data of the target user to each type of advertisement content;

[0068] This step measures user receptivity to various types of advertising content by calculating the repeat click rate and average click interval based on the target user's historical click data. The repeat click rate reflects the proportion of users who repeatedly click on the same type of ad, while the average click interval reflects the time interval between multiple clicks. Both metrics provide different insights into the persistence of user interest in this type of ad. A higher repeat click rate and a shorter average click interval indicate a higher likelihood of user interest in this type of ad content.

[0069] In specific implementation, the system can use a big data processing engine to perform real-time or offline calculations on massive amounts of historical click data. First, the click data is aggregated according to the category of advertising content, and then the repeat click rate and average click interval are calculated for each type of advertisement. The definition of repeat clicks can be flexibly set according to the advertising recommendation scenario, such as multiple clicks in a single ad delivery, multiple clicks in a day, or multiple clicks in a month. For the calculation of the average click interval, it is necessary to comprehensively consider the timestamps of multiple click behaviors and divide the time difference between the earliest and latest clicks by the number of clicks minus one.

[0070] S103: Determine the consumption conversion links of target users in different time periods based on the purchase behavior data;

[0071] Based on purchasing behavior data, the system determines the consumption conversion link of target users in different time periods. The consumption conversion link includes the sequence of advertising content from the first click to the final purchase.

[0072] Purchasing behavior data reflects users' final purchasing decisions. Analyzing this data helps us gain insight into their shopping habits and psychology. This step leverages target users' purchasing behavior data to identify their consumer conversion chain across different time periods. The consumer conversion chain refers to the entire process from a user's initial ad exposure to a purchase, encompassing the exposure and influence of a series of advertising content. Research shows that user purchasing decisions are typically not made overnight but are instead influenced by a comprehensive series of advertisements. Understanding the composition and timing of this series of advertising content is crucial for optimizing advertising strategies.

[0073] In specific implementation, the system first needs to correlate purchase behavior data with ad click data to identify advertising content that users may have been exposed to prior to each purchase. This requires considering the time difference between purchase and click behavior, as well as the relevance of the advertising content to the purchased product. After determining the series of advertising content corresponding to each purchase, the system can sort these advertising content by time to obtain a series of consumer conversion links. The system can compare the similarities and differences between these links for different time periods, analyze whether there are differences in users' shopping decisions during different time periods, and discover the underlying patterns.

[0074] One possible technical problem is that in actual applications, not every purchase behavior can be directly associated with the previous advertisement click. Purchase behavior is influenced by many factors, and the user may not order immediately after seeing the advertisement. One solution is that the system can use machine learning methods such as probabilistic graphical models to automatically learn the correlation strength and time decay law between different advertisement contents and purchase behaviors from a large amount of historical data, and flexibly match behaviors that are difficult to directly associate. At the same time, business rules and expert experience can be used to constrain and optimize the construction of conversion links.

[0075] S104、According to the consumption conversion link, determine the key advertisement content combination and the optimal display timing;

[0076] The system determines the key advertisement content combination and the optimal display timing according to the consumption conversion link. Specifically, the time distribution of each type of advertisement content triggering the target user's click behavior in the consumption conversion link is counted to obtain time sensitivity data.

[0077] Based on the time sensitivity data, the conversion probability of different advertisement content combinations within a preset time window is calculated, and the conversion probability represents the probability of the advertisement content combination triggering the target user's purchase behavior.

[0078] Select the advertisement content combination with a conversion probability greater than a preset probability threshold as the key advertisement content combination.

[0079] Analyze the optimal display time interval of each advertisement content in the key advertisement content combination, establish a display rule based on the display rule, and generate an optimal display timing based on the display rule.

[0080] After determining the user's consumption conversion link, this step further analyzes these links to determine the key advertisement content combination that has the greatest impact on the user's purchase decision and the optimal display timing of these key contents. The so-called key advertisement content combination refers to some advertisement contents that frequently co-occur in the consumption conversion link and have a high correlation with the final purchase behavior. The combination of these contents can effectively trigger the user's purchase desire, so these contents should be given priority in subsequent advertisement placement. The optimal display timing refers to an optimal display order and time interval between the key advertisement contents mined, so that the influence of these advertisement contents can be maximized.

[0081] To mine the key ad content combinations, the system can use frequent pattern mining, association rule mining, and other algorithms to automatically discover frequently co-occurring ad content combinations from a large number of consumption conversion links and evaluate their association strength with purchase behavior. For calculating the optimal display timing, it can be modeled as a sequence optimization problem, with the goal of maximizing the probability that the ad content sequence triggers purchase behavior. This can be solved with the help of sequence models such as Hidden Markov Models, Recurrent Neural Networks, etc. At the same time, the attribute characteristics of different ad contents and the user's historical behavior also need to be considered, and the sequence optimization problem needs to be solved individually.

[0082] S105, filtering out ad contents with a similarity greater than a preset first similarity threshold and a click frequency less than a preset frequency from the preset ad library to obtain an ad content exploration pool;

[0083] The system filters out new ad contents with exploration value in the ad library, providing more candidates for subsequent ad sequence optimization. Specifically, the system calculates the similarity between each ad content in the ad library and the known key ad content combinations, and selects contents with a similarity higher than a certain threshold. At the same time, in order to avoid selecting ads that users have already become aesthetically fatigued of, the historical click frequency of the candidate contents is also limited. Ad contents that meet both conditions are added to the ad content exploration pool and become candidates for subsequent dynamic insertion of ad sequences. This approach can explore some new contents that are similar to known key ads and have potential, while also controlling the breadth and depth of exploration, balancing development and utilization.

[0084] To calculate the similarity between ad contents, natural language processing and computer vision techniques can be used. For example, for text-based ads, topic models, Word2Vec, and other algorithms can be used to calculate the semantic similarity of the text; for image-based ads, convolutional neural networks can be used to extract high-level semantic features of the image and then calculate the similarity between the features. For rich media ads of complex types, the similarity of different modalities such as text, images, and videos needs to be integrated. In addition, the system can also combine the metadata information of the ad content (such as the advertiser, the channel, the audience targeting conditions, etc.) to optimize the similarity calculation.

[0085] S106, inserting the ad contents in the ad content exploration pool into the optimal display timing in a progressive manner to form a dynamic ad placement sequence;

[0086] The system inserts the ad contents in the ad content exploration pool into the optimal display timing in a progressive manner to form a dynamic ad placement sequence. Specifically, the content similarity between each ad content in the ad content exploration pool and each ad content in the key ad content combination is calculated.

[0087] Grouping the advertising contents in the advertising content exploration pool according to content similarity to obtain similar advertising groups, each similar advertising group corresponding to one advertising content in the key advertising content combination;

[0088] In each group of similar advertisements, the advertisement contents are sequentially inserted after the display timing point of the corresponding key advertisement contents in the order of content similarity from high to low.

[0089] In the previous steps, the system has already established a relatively stable sequence of key advertising content and a pool of new advertising content to explore. This step aims to dynamically and progressively insert the new advertising content from the exploration pool into the existing optimal display sequence, thereby forming a dynamic, personalized advertising delivery sequence. Progressive insertion means that the addition of new advertising content is gradual and controllable, leveraging the exploration pool to enrich delivery content while minimizing the impact on the existing optimal sequence structure. This ensures that the overall effectiveness of advertising delivery is maximized while exploring.

[0090] In specific implementation, the system must first map and associate the new advertising content in the advertising content exploration pool with the original content in the key advertising content sequence. One approach is to calculate the similarity matrix between the new and old advertising content to find the most similar insertion position of each new advertising content in the original sequence. It can also consider the contextual relevance and complementarity of the new advertising content with the original sequence, as well as the structural impact of the new advertisement on the original sequence after insertion. After determining the insertion position of each new advertisement, the system can control the actual insertion probability and frequency of the new advertisement to achieve progressive exploration. For example, an initial insertion probability can be set, and its insertion probability can be dynamically adjusted as the effectiveness of the new advertising content is observed.

[0091] A key issue in this step is how to strike a balance between exploring new ad content and protecting the original sequence structure, promoting the discovery of new content while preventing the destruction of the original sequence structure due to excessive insertion. One approach is for the system to establish a sequence structure evaluation model that measures the difference between the sequence structure after the insertion of the new ad and the original sequence. When inserting new ads, this difference is strictly controlled within a certain threshold. The system can also track the performance indicators of new and old ad content in real time. Once it finds that the new ad content is performing poorly or better than the original content, its insertion strategy is dynamically adjusted to reduce or expand the presentation of the new content. This is actually an online parameter adjustment and optimization process that requires balancing content diversity and delivery stability.

[0092] S107: delivering the final advertisement content corresponding to the dynamic advertisement delivery sequence to the target user according to the dynamic advertisement delivery sequence.

[0093] After the calculation and optimization of the previous steps, the system finally gets a dynamically generated and personalized advertising sequence. The task of this step is to really put the actual advertising content corresponding to the sequence into the target users according to the optimized time sequence and display strategy. This is the last link of advertising optimization, which directly affects the actual observation experience of users and the monetization effect of the platform. The channel, timing, frequency, and material style of the advertising need to be strictly set according to the dynamic advertising sequence obtained by the previous optimization.

[0094] In specific implementation, the system needs to be connected with the advertising delivery platform (such as the advertising alliance, DSP, etc.), and the specific advertising content corresponding to each advertising position in the dynamic sequence is sent to the delivery platform through the API interface. This requires that the material mapping and optimization at the advertising creative level have been completed when the dynamic sequence is generated. The system also needs to reasonably control the rhythm of the delivery, both to ensure the exposure rate of key advertising content and to avoid too dense advertising bombardment that causes user aversion. The time interval of the delivery can be dynamically adjusted according to the time mark and content weight of the sequence. At the same time, the quality of the delivery channel also needs to be monitored to eliminate some false traffic and invalid exposure, and to improve the cost performance of advertising delivery.

[0095] In the above embodiment, the historical click data and purchase behavior data of the target user are obtained, the acceptance degree of the user to the advertising content is calculated based on the repeated click rate and the average click interval, and the advertising preference characteristics of the user can be grasped. The key advertising content combination and the optimal display time sequence are determined based on the consumption conversion link of the user, and the advertising content sequence that can most effectively trigger the user's purchase behavior can be identified. The advertising content with high similarity but low display frequency is selected from the preset advertising library to form an exploration pool, and it is inserted into the optimal display time sequence in a progressive manner, which not only maintains the effectiveness of the advertising delivery, but also expands the range of the user's advertising experience, so that the advertising delivery conforms to the user's acceptance habit and maintains the freshness of the content, and on the basis of maintaining the existing delivery effect, the delivery effect is continuously enhanced.

[0096] To realize the advertising delivery method based on big data analysis provided in the embodiments of the present application, the system needs to monitor and dynamically adjust the delivery effect in real time after completing the execution of the advertising delivery sequence. Therefore, on the basis of the above embodiment, the present application also provides an advertising content optimization method based on feedback data, which realizes the continuous optimization of the advertising delivery effect by establishing a dynamic evaluation and adjustment mechanism for the advertising content. The optimization method and the above advertising delivery method complement each other, and together constitute a complete advertising delivery optimization system. The following will be described in combination with Figure 2 An advertising content optimization method based on feedback data in the embodiments of the present application is described as follows:

[0097] Please refer toFigure 2 , as a flowchart of one embodiment of the advertisement content optimization method based on feedback data in the present application.

[0098] S201, real-time collection of feedback data of target users, and calculation of user engagement score of final advertisement content based on feedback data;

[0099] The system collects the feedback data of the target users in real time, and calculates the user engagement score of the final advertisement content based on the feedback data, wherein the feedback data includes click behavior, dwell time and interactive operation. Specifically, the click behavior, dwell time and interactive operation are respectively assigned a preset weight coefficient;

[0100] According to the feedback data, a comprehensive engagement index is calculated, which is the weighted sum of the click behavior, dwell time and interactive operation and the corresponding weight coefficient;

[0101] Based on the comprehensive engagement index, a time decay coefficient of each final advertisement content is calculated;

[0102] The comprehensive engagement index is multiplied by the time decay coefficient to obtain the user engagement score.

[0103] In this step, the system evaluates the actual effect of the advertisement content by real-time collection of feedback data generated by the target users when browsing the advertisement content. The feedback data mainly includes click behavior, dwell time and interactive operation, etc. which can comprehensively reflect the interest degree and engagement of users to the advertisement content. The system takes these feedback data as the basis for calculating the user engagement score, and obtains the scoring result reflecting the quality and attractiveness of the advertisement content by quantitative analysis of user behavior.

[0104] In specific implementation, the system can record the user's click, dwell and interactive behavior in real time by embedding a data collection module in the advertisement display page. The collected raw data can be transmitted to the background server for analysis and calculation. For different types of feedback data, the system can set corresponding weight coefficients to reflect the importance of each index. For example, a higher weight can be given to click behavior, while relatively lower weights can be given to dwell time and interactive operation. Through weighted calculation of feedback data and weight coefficients, a comprehensive engagement index is obtained. On this basis, the system can also introduce a time decay factor in combination with the display time of the advertisement content, to reduce the engagement score of the advertisement content published earlier, and pay more attention to the influence of recent feedback data. Finally, the comprehensive engagement index is multiplied by the time decay coefficient to obtain the dynamically updated user engagement score.

[0105] S202, moving the advertisement content whose user engagement score is lower than the preset engagement threshold out of the advertisement content exploration pool;

[0106] In this step, the system filters and optimizes the advertising content based on the calculated user engagement scores. By setting a pre-set engagement threshold, the system can automatically identify advertising content with low engagement and insufficient appeal, and remove it from the advertising content exploration pool. This process can achieve dynamic updating and optimization of advertising content, ensuring that high-quality, high-engagement advertising content is always retained in the exploration pool, improving overall delivery effectiveness.

[0107] In specific implementation, the system can set a reasonable pre-set engagement threshold based on historical data and business needs. This threshold can be a fixed value or a dynamically adjusted range. When the user engagement score of a certain advertising content is below this threshold, the system can automatically mark it as low-quality content and remove it from the exploration pool. At the same time, the system can also record relevant information of the removed content, such as content theme, creative elements, etc., for subsequent analysis and optimization.

[0108] S203, cluster analysis is performed on the remaining advertising content in the advertising content exploration pool to obtain subsets of advertising content of different theme categories;

[0109] In this step, the system further analyzes and processes the optimized advertising content exploration pool. Through clustering algorithms, the system can automatically classify advertising content with similar themes or features into the same subset, forming combinations of advertising content of different theme categories. This process can help the system better understand and grasp the distribution characteristics of advertising content, providing support for subsequent content optimization and recommendation.

[0110] In specific implementation, the system can select appropriate clustering algorithms, such as K-means clustering, hierarchical clustering, etc., to automatically group advertising content. Clustering algorithms can calculate the similarity between content based on multi-dimensional information such as text features, image features, video features, etc., and classify content into different clusters according to the similarity size. Each cluster represents a subset of advertising content of a theme category. The system can visualize the clustering results to intuitively analyze the content distribution of different theme categories.

[0111] In the implementation of this step, the clustering results may not be accurate or practical enough, affecting the subsequent content optimization effect. To solve this problem, the system can introduce manual intervention and feedback mechanism, allowing business experts to review and adjust the clustering results. Experts can reclassify certain boundary samples or outliers based on their experience and judgment, improving the accuracy of clustering. At the same time, the system can also use expert feedback to iteratively optimize the clustering algorithm, dynamically adjusting clustering parameters and strategies to better adapt to actual business scenarios. In addition, the system can regularly evaluate and update the clustering results, adjust the division of theme categories in a timely manner according to the changes of advertising content, and ensure the timeliness and accuracy of the clustering results.

[0112] S204, calculate the number proportion of each advertising content subset, and compare it with the preset balanced distribution proportion;

[0113] The system calculates the number proportion of each advertising content subset and compares it with the preset balanced distribution proportion, which includes the target proportion corresponding to each advertising content subset.

[0114] In this step, after completing the clustering of advertising content, the system further analyzes the number distribution of each theme category subset. By calculating the proportion of the number of advertising content in each subset to the total, the system can comprehensively understand whether the content distribution of different theme categories is balanced. At the same time, the system also needs to compare the actual distribution with the preset balanced distribution proportion to find possible content deviations or deficiencies.

[0115] In specific implementation, the system can count the number of advertising content in each theme category subset based on the clustering results and calculate its percentage of the total content. The preset balanced distribution proportion can be set according to business needs and user preferences, for example, requiring the content number proportion of each theme category to be close, or requiring the content proportion of certain popular theme categories to be slightly higher than other categories. The system compares the actual distribution calculated with the preset proportion to calculate the difference between the two.

[0116] S205, supplement advertising content with a number proportion less than the corresponding target proportion from the preset advertising library.

[0117] The system supplements advertising content with a number proportion less than the corresponding target proportion from the preset advertising library. Specifically, the advertising content with a number proportion less than the corresponding target proportion is determined as the advertising content to be supplemented;

[0118] Calculate the difference between the number proportion of the advertising content to be supplemented and the corresponding target proportion;

[0119] Match the candidate advertising content corresponding to the advertising content to be supplemented in the preset advertising library;

[0120] calculating the similarity of the candidate advertisement content and the existing content in the corresponding subset of advertisement content;

[0121] selecting the candidate advertisement content with a similarity less than a preset second similarity threshold as the supplementary content, and adding the supplementary content to the advertisement content exploration pool, the number of the supplementary content being the same as the difference.

[0122] In this step, the system selects relevant content from the preset advertisement library to supplement for the topic categories with unbalanced quantity distribution. Through content supplementation, the system can realize dynamic expansion and optimization of the advertisement content exploration pool, ensuring that the content quantity of each topic category meets the preset balanced distribution requirement. At the same time, content supplementation also helps to introduce new creative elements and forms, improving the diversity and appeal of the advertisement content.

[0123] In specific implementation, the system first needs to identify the topic categories with a quantity proportion less than the corresponding target proportion, and calculate the size of the content gap. Then, the system can retrieve candidate content related or similar to the topic categories in the preset advertisement library according to the characteristics of the topic categories. The candidate content can be matched and sorted based on elements such as title, keyword, and text description. In order to avoid content repetition or excessive similarity, the system also needs to calculate the similarity of the candidate content and the existing content in the exploration pool, and eliminate the content with a similarity exceeding a preset threshold. Finally, the system selects content with a quantity the same as the size of the gap from the candidate content with a high ranking and a low similarity, and supplements it to the corresponding subset of topic categories.

[0124] In the above embodiment, real-time feedback data of users on the launched advertisement is collected, including click behavior, dwell time, and interactive operation, the user engagement score of the advertisement content is calculated, and the advertisement content with a user engagement score lower than a preset engagement threshold is removed from the exploration pool, so that the advertisement launching system can dynamically adjust the quality level of the advertisement content library and continuously optimize the advertisement content exploration pool. By timely cleaning up the advertisement content with low user engagement, it is ensured that only high-quality advertisement content that can effectively attract user attention and interaction is retained in the exploration pool, improving the overall quality and effect of advertisement launching. Cluster analysis can be performed on the remaining advertisement content in the advertisement content exploration pool to find the distribution of advertisement content of different topic categories. By calculating the quantity proportion of each subset of advertisement content and comparing it with the preset balanced distribution ratio, the unbalanced distribution of advertisement content can be identified. According to the comparison result of the actual proportion and the target proportion of each category of advertisement content, the advertisement content with insufficient quantity can be supplemented from the preset advertisement library, so that the reasonable distribution of different topic categories in the advertisement content exploration pool can be maintained, the diversity of advertisement content is maintained, and the situation of excessive concentration or serious shortage of advertisement content of certain topic categories is avoided, so that the advertisement launching can cover a wider range of user interest points, improving the overall advertisement launching effect.

[0125] The system in the embodiments of the present application is described from the perspective of hardware processing below. Please refer to Figure 3 FIG. 1 is a schematic diagram of an entity device structure of an advertisement delivery system based on big data analysis provided by the embodiments of the present application.

[0126] It should be noted that Figure 3 The structure of the system shown is only an example and should not bring any limitation to the functions and use range of the embodiments of the present application.

[0127] As Figure 3 shown, the system includes a central processing unit (CPU) 301, which can perform various appropriate actions and processes according to programs stored in a read-only memory (ROM) 302 or loaded from a storage portion 308 into a random access memory (RAM) 303, such as performing the methods in the above embodiments. In the RAM 303, various programs and data required for system operation are also stored. The CPU 301, the ROM 302, and the RAM 303 are connected to each other through a bus 304. An input / output (I / O) interface 305 is also connected to the bus 304.

[0128] The following components are connected to the I / O interface 305: an input portion 306 including a camera, an infrared sensor, and the like; an output portion 307 including a liquid crystal display (LCD), a speaker, and the like; a storage portion 308 including a hard disk and the like; and a communication portion 309 including a network interface card such as a LAN (Local Area Network) card, a modem, and the like. The communication portion 309 performs communication processing via a network such as the Internet. A drive 310 is also connected to the I / O interface 305 as needed. A removable medium 311 such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, and the like is mounted on the drive 310 as needed, so that a computer program read therefrom is installed in the storage portion 308 as needed.

[0129] In particular, the processes described above with reference to the flowcharts can be implemented as a computer software program in accordance with embodiments of the present application. For example, embodiments of the present application include a computer program product comprising a computer program carried on a computer readable medium, the computer program comprising computer programs for executing the methods illustrated by the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via the communication section 309, and / or installed from the removable medium 311. When the computer program is executed by the central processing unit (CPU) 301, various functions defined in the present application are executed.

[0130] It should be noted that the computer readable medium shown in the embodiments of the present application can be a computer readable signal medium or a computer readable storage medium or any combination of the two. The computer readable storage medium may, for example, be but is not limited to an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus or device, or any combination of the above. More specific examples of the computer readable storage medium can include, but are not limited to, an electrical connection having one or more wires, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM), a flash memory, an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present application, the computer readable storage medium can be any tangible medium that contains or stores a program that can be used by or in connection with an instruction execution system, apparatus or device. In the present application, the computer readable signal medium can include a data signal carried in a baseband or as part of a carrier wave, in which a computer readable program is carried. Such a propagated data signal can take a variety of forms, including but not limited to an electromagnetic signal, an optical signal or any suitable combination of the above.

[0131] The computer program product of the present application can be a computer program including a plurality of program instructions. The plurality of program instructions can include one or more of program instructions for implementing the functions of the system, the method, and the computer program product according to the embodiments of the present application. The computer program product of the present application can also be a computer program including a plurality of program instructions. The plurality of program instructions can include one or more of program instructions for implementing the functions of the system, the method, and the computer program product according to the embodiments of the present application.

[0132] As another aspect, the present application also provides a computer readable storage medium, which can be included in the system described in the above embodiments, or can exist independently without being assembled into the system. The above storage medium carries one or more computer programs, which, when executed by a processor of a system, enable the system to implement the method provided in the above embodiments.

[0133] The above embodiments are only used to illustrate the technical solutions of the present application, but not limit the present application; even though the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: they can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacements to some technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present application.

[0134] In the above embodiments, according to the context, the term "when" can be interpreted as "if" or "after" or "in response to determining" or "in response to detecting". Similarly, according to the context, the phrase "upon determining" or "if detecting (the stated condition or event)" can be interpreted as "if determining" or "in response to determining" or "upon detecting (the stated condition or event)" or "in response to detecting (the stated condition or event)".

[0135] In the above embodiments, all or part of the embodiments can be implemented by software, hardware, firmware or any combination thereof. When implemented by software, all or part of the embodiments can be implemented in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of the present application are generated. The computer can be a general purpose computer, a special purpose computer, a computer network, or other programmable apparatus. The computer instructions can be stored in a computer readable storage medium or transmitted from one computer readable storage medium to another computer readable storage medium, for example, the computer instructions can be transmitted from one website, computer, server or data center to another website, computer, server or data center through wired (such as coaxial cable, optical fiber, digital subscriber line) or wireless (such as infrared, wireless, microwave, etc.) manner. The computer readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server, data center, etc. integrated with one or more available media. The available media can be a magnetic medium (for example, a floppy disk, a hard disk, a magnetic tape), an optical medium (for example, a DVD), or a semiconductor medium (for example, a solid state disk) and the like.

[0136] Those of ordinary skill in the art understand that all or part of the processes in the above embodiments can be implemented by a computer program to instruct the relevant hardware, which can be stored in a computer readable storage medium. The program can include the processes of the above method embodiments when executed. The aforementioned storage medium includes ROM or random access memory (RAM), magnetic disk or optical disk, and various media that can store program codes.

Claims

1. An advertising delivery method based on big data analysis, characterized in that: include: Obtain the target user's historical click data and purchase behavior data, the historical click data including the clicked ad content, click time, and click frequency, and the purchase behavior data including the purchased product information, purchase time, and purchase amount; Determining, based on the purchase behavior data, a consumption conversion link of the target user in different time periods, wherein the consumption conversion link includes an advertisement content sequence from an initial click to a final purchase; Determine the key advertising content combination and optimal display timing based on the consumption conversion link, specifically including: Counting the time distribution of the target user's click behavior triggered by various types of advertising content in the consumption conversion link to obtain time sensitivity data; Based on the time sensitivity data, calculating the conversion probability of different advertising content combinations within a preset time window, the conversion probability representing the probability that the advertising content combination triggers the target user's purchase behavior; Selecting the advertising content combination with the conversion probability greater than a preset probability threshold as the key advertising content combination; Analyzing the optimal display time interval of each advertisement content in the key advertisement content combination, establishing a display rule, and generating an optimal display timing based on the display rule; Screening out advertising content from a preset advertising library, whose similarity to the key advertising content combination is greater than a preset first similarity threshold and whose click frequency is less than a preset frequency, to obtain an advertising content exploration pool; Inserting the advertising content in the advertising content discovery pool into the optimal presentation sequence in a progressive manner to form a dynamic advertising delivery sequence; The final advertisement content corresponding to the dynamic advertisement delivery sequence is delivered to the target user according to the dynamic advertisement delivery sequence.

2. The method according to claim 1, characterized in that The step of inserting the advertisement contents in the advertisement content discovery pool into the optimal presentation sequence in a progressive manner specifically includes: Calculating content similarity between each advertisement content in the advertisement content exploration pool and each advertisement content in the key advertisement content combination; Grouping the advertising contents in the advertising content exploration pool according to the content similarity to obtain similar advertising groups, each of the similar advertising groups corresponding to one advertising content in the key advertising content combination; In each of the similar advertisement groups, the advertisement contents are sequentially inserted after the presentation timing point of the corresponding key advertisement contents in the order of the content similarity from high to low.

3. The method according to claim 1, characterized in that After delivering the final advertisement content corresponding to the dynamic advertisement delivery sequence to the target user according to the dynamic advertisement delivery sequence, the method further includes: Collecting feedback data from the target user in real time, including click behavior, stay time, and interactive operations; Calculating a user engagement score for the final advertising content based on the feedback data; The advertising content with the user engagement score lower than a preset engagement threshold is removed from the advertising content exploration pool.

4. The method according to claim 3, characterized in that Calculating the user engagement score of the final advertising content based on the feedback data specifically includes: Assigning preset weight coefficients to the click behavior, the stay duration, and the interactive operation respectively; Calculating a comprehensive engagement index based on the feedback data, where the comprehensive engagement index is a weighted sum of the click behavior, the dwell time, and the interactive operation and corresponding weight coefficients; Calculating a time decay coefficient of each of the final advertising contents based on the comprehensive engagement index; The comprehensive engagement index is multiplied by the time decay coefficient to obtain a user engagement score.

5. The method according to claim 3, characterized in that After removing the advertising content having the user engagement score lower than the preset engagement threshold from the advertising content exploration pool, the method further includes: Performing cluster analysis on the remaining advertising content in the advertising content exploration pool to obtain advertising content subsets of different subject categories; Calculating the quantity ratio of each of the advertising content subsets and comparing it with a preset balanced distribution ratio, wherein the preset balanced distribution ratio includes a target ratio corresponding to each of the advertising content subsets; Advertisement contents whose proportion is less than the corresponding target proportion are supplemented from the preset advertisement library.

6. The method according to claim 5, characterized in that The step of supplementing the advertising content whose proportion is less than the corresponding target proportion from the preset advertising library specifically includes: Determine the advertising content whose proportion is less than the corresponding target proportion as the advertising content to be supplemented; Calculating the difference between the proportion of the quantity of the to-be-supplemented advertising content and the corresponding target proportion; Matching candidate advertisement content corresponding to the advertisement content to be supplemented in the preset advertisement library; Calculating similarity between the candidate advertisement content and existing content in the corresponding advertisement content subset; The candidate advertising content with the similarity less than a preset second similarity threshold is selected as supplementary content, and the supplementary content is added to the advertising content exploration pool. The number of the supplementary content is the same as the difference.

7. An advertising delivery system based on big data analysis, characterized in that: The system comprises: One or more processors and a memory; the memory is coupled to the one or more processors, the memory is used to store computer program code, the computer program code includes computer instructions, and the one or more processors call the computer instructions to cause the system to execute the method according to any one of claims 1 to 6.

8. A computer-readable storage medium comprising instructions, characterized in that: When the instructions are executed on a system, the system is caused to perform the method according to any one of claims 1 to 6.

9. A computer program product, characterized in that When the computer program product is run on a system, the system is caused to perform the method according to any one of claims 1 to 6.

Images