DDA-based advertisement multi-touch effect feedback method and system

Abstract

The application relates to the technical field of advertisement effect feedback, and particularly discloses an advertisement multi-touch point effect feedback method and system based on DDA, which comprises the following steps: fatigue coefficients are calculated based on a click rate decay rate, a closing rate and the exposed number of times of a current user; a video complete play rate is acquired, a content coefficient is calculated based on the video complete play rate, and a tolerance coefficient is calculated according to the fatigue coefficient and the content coefficient; a value coefficient is calculated according to a page stay duration and a scroll depth; and a target display frequency upper limit is calculated based on a preset basic frequency, the tolerance coefficient and the value coefficient. Through the fusion of multi-touch point data such as user behaviors, content quality and page scenes, the application dynamically evaluates user states and advertisement values, forms a feedback closed loop of real-time perception, calculation and adjustment, overcomes the defects of traditional methods that rely on static rules and have single feedback, and obtains extremely comprehensive and accurate advertisement feedback effects, which equivalently improves user experience.

Description

Technical Field

[0001] This invention relates to the field of advertising effectiveness feedback technology, specifically a method and system for providing multi-touchpoint advertising effectiveness feedback based on DDA (Directed Marketing). Background Technology

[0002] In the digital advertising ecosystem, the refined management of ad display frequency is directly related to the balance between advertisers' ROI and end-user experience. Traditional frequency control strategies generally rely on static rules, such as setting a uniform daily exposure cap for all users, or mechanically distributing ad placements based on the total budget of the advertising plan. While such solutions are convenient to implement, they have fundamental flaws: they cannot perceive the real-time behavioral differences of individual users, ignore the dynamic impact of ad creative content quality, and cannot adapt to the contextual value of different page scenarios. Therefore, how to provide a multi-source information perception solution to obtain more comprehensive advertising performance feedback information is the technical problem that this invention aims to solve. Summary of the Invention

[0003] The purpose of this invention is to provide a method and system for multi-touchpoint advertising effect feedback based on DDA, so as to solve the problems mentioned in the background art.

[0004] To achieve the above objectives, the present invention provides the following technical solution:

[0005] A method for providing multi-touchpoint advertising performance feedback based on DDA, the method comprising:

[0006] Get the click-through rate of the ad, calculate the click-through rate decay rate, and calculate the fatigue coefficient based on the click-through rate decay rate, the close rate, and the number of times the ad has been exposed to the current user;

[0007] Obtain the video completion rate, calculate the content coefficient based on the video completion rate, and calculate the acquisition tolerance coefficient based on the fatigue coefficient and the content coefficient;

[0008] Obtain the page dwell time and scroll depth, and calculate the value coefficient based on the page dwell time and scroll depth;

[0009] Based on the preset base frequency, tolerance coefficient, and value coefficient, calculate the target display frequency limit and adjust the current display frequency limit to the target display frequency limit.

[0010] As a further aspect of the present invention: the steps of obtaining the click-through rate of the advertisement, calculating the click-through rate decay rate, and calculating the fatigue coefficient based on the click-through rate decay rate, the close rate, and the number of times the creative has been exposed to the current user include:

[0011] Get the click-through rate (CTR) of the ads and calculate the CTR decay rate;

[0012] Get the close rate and the number of times the creative has been exposed to the current user;

[0013] The current click-through rate decay rate is processed by maximum-min normalization to obtain the click-through rate decay rate index.

[0014] The current closure rate is compared with a preset threshold, and the closure rate index is obtained by using the min function to limit the upper limit of the ratio to 1.

[0015] The creative exposure saturation is obtained by comparing the number of times the creative has been exposed to the current user with the maximum number of times the creative's freshness has been reached.

[0016] Based on the click-through rate decay rate index, the close rate index, and the creative exposure saturation, the three factors are weighted and summed using preset weighting coefficients to obtain the fatigue coefficient. The sum of each weighting coefficient is 1, and the fatigue coefficient ranges from 0 to 1, reflecting the degree of user fatigue with the advertisement.

[0017] As a further aspect of the present invention: the step of obtaining the click-through rate of the advertisement and calculating the click-through rate decay rate includes:

[0018] Get the current click-through rate and the baseline click-through rate;

[0019] The current click-through rate is compared with the baseline click-through rate, and the complement of the ratio is taken as the click-through rate decay rate.

[0020] As a further aspect of the present invention: the steps of obtaining the video completion rate, calculating the content coefficient based on the video completion rate, and calculating and obtaining the tolerance coefficient according to the fatigue coefficient and the content coefficient include:

[0021] Get the video completion rate;

[0022] Based on the video completion rate, it is converted into content coefficients through a non-linear mapping function. The non-linear mapping function is centered on the benchmark completion rate threshold, and the steepness of the mapping curve is controlled by the slope parameter. The output range is between 0 and 1, reflecting the attractiveness of the advertising content.

[0023] Substitute the fatigue coefficient and content coefficient into the formula. Obtain tolerance coefficient Tolerance coefficient Output range 0-1, tolerance coefficient A larger value indicates a higher tolerance level, and the frequency can be adjusted to maximize value. The fatigue coefficient, For content coefficients, This is a content compensation factor that controls the degree to which content quality compensates for fatigue.

[0024] As a further aspect of the present invention: the step of obtaining the page dwell time and scroll depth, and calculating the value coefficient based on the page dwell time and scroll depth includes:

[0025] Get the page dwell time and scroll depth;

[0026] The page dwell time is normalized by the maximum and minimum, and then limited to 0-1 to obtain the page dwell time index.

[0027] The current scroll depth is processed and its maximum value is compared to obtain the scroll depth index;

[0028] Based on the page dwell time index and scroll depth index, the two are weighted and combined using configurable weight parameters, and the results are normalized to obtain a value coefficient ranging from 0 to 1. This coefficient reflects the depth of user interaction on the current page.

[0029] As a further aspect of the present invention: the step of calculating the target display frequency upper limit based on a preset base frequency, tolerance coefficient, and value coefficient, and adjusting the current display frequency upper limit to the target display frequency upper limit includes:

[0030] Obtain the basic frequency, tolerance coefficient, and value coefficient;

[0031] Based on the base frequency, adjustments are made using the tolerance coefficient, and the gain is adjusted based on the value coefficient to obtain the upper limit of the target display frequency.

[0032] The present invention also provides a DDA-based multi-touchpoint advertising performance feedback system, the system comprising:

[0033] The fatigue level determination module is used to obtain the click-through rate of the advertisement, calculate the click-through rate decay rate, and calculate the fatigue coefficient based on the click-through rate decay rate, the close rate, and the number of times the creative has been exposed to the current user.

[0034] The tolerance assessment module is used to obtain the video completion rate, calculate the content coefficient based on the video completion rate, and calculate and obtain the tolerance coefficient based on the fatigue coefficient and the content coefficient.

[0035] The value assessment module is used to obtain page dwell time and scroll depth, and calculate the value coefficient based on page dwell time and scroll depth;

[0036] The ad frequency adjustment module is used to calculate the target display frequency limit based on preset base frequency, tolerance coefficient, and value coefficient, and adjust the current display frequency limit to the target display frequency limit.

[0037] As a further aspect of the present invention: the fatigue determination module includes:

[0038] The click-through rate analysis unit is used to obtain the click-through rate of advertisements and calculate the click-through rate decay rate;

[0039] The parameter acquisition unit is used to obtain the close rate and the number of times the creative has been exposed to the current user;

[0040] The decay index processing unit is used to perform maximum-min normalization on the current click-through rate decay rate to obtain the click-through rate decay rate index.

[0041] The closing index processing unit is used to process the ratio of the current closing rate to a preset threshold, and then use the min function to limit the upper limit of the ratio to 1 to obtain the closing rate index.

[0042] The saturation output unit is used to process the ratio of the number of times the creative has been exposed to the current user to the maximum number of times the creative's freshness has been reached, and obtain the creative's exposure saturation.

[0043] The weighted summation unit is used to sum the click-through rate decay rate index, the close rate index, and the creative exposure saturation using preset weighting coefficients to obtain a fatigue coefficient. The sum of each weighting coefficient is 1, and the fatigue coefficient ranges from 0 to 1, reflecting the degree of user fatigue with the advertisement.

[0044] As a further aspect of the present invention: the tolerance assessment module includes:

[0045] The completion rate acquisition unit is used to obtain the video completion rate;

[0046] The content analysis unit is used to convert video completion rate into content coefficients through a non-linear mapping function. The non-linear mapping function is centered on the benchmark completion rate threshold, and the steepness of the mapping curve is controlled by the slope parameter. The output range is between 0 and 1, reflecting the attractiveness of the advertising content.

[0047] The tolerance coefficient generation unit is used to substitute the fatigue coefficient and content coefficient into the formula. Obtain tolerance coefficient Tolerance coefficient Output range 0-1, tolerance coefficient A larger value indicates a higher tolerance level, and the frequency can be adjusted to maximize value. The fatigue coefficient, For content coefficients, This is a content compensation factor that controls the degree to which content quality compensates for fatigue.

[0048] As a further aspect of the present invention: the value assessment module includes:

[0049] The page information acquisition unit is used to obtain the page dwell time and scroll depth;

[0050] The dwell time index acquisition unit is used to perform maximum-minimum normalization on the page dwell time, and after limiting the range to 0-1, obtain the page dwell time index.

[0051] The scrolling index determination unit is used to process the current scrolling depth with its maximum value to obtain the scrolling depth index;

[0052] The value coefficient generation unit is used to perform weighted comprehensive processing on the page dwell time index and scroll depth index through configurable weight parameters, and normalize the result to obtain a value coefficient between 0 and 1. This coefficient reflects the user's interaction depth on the current page.

[0053] Compared with existing technologies, the beneficial effects of this invention are: by integrating multi-touchpoint data such as user behavior, content quality, and page scenarios, this invention dynamically evaluates user status and advertising value, forming a feedback loop of real-time perception, calculation, and adjustment. This overcomes the shortcomings of traditional methods that rely on static rules and have single feedback, resulting in extremely comprehensive and accurate advertising feedback, thereby indirectly improving the user experience. Attached Figure Description

[0054] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention.

[0055] Figure 1 This is a flowchart of a DDA-based multi-touchpoint advertising performance feedback method.

[0056] Figure 2 This is a block diagram of the composition structure of a DDA-based multi-touchpoint advertising performance feedback system. Detailed Implementation

[0057] To make the technical problems to be solved, the technical solutions, and the beneficial effects of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present invention and are not intended to limit the present invention.

[0058] Figure 1 This is a flowchart of a DDA-based multi-touchpoint advertising performance feedback method. In this embodiment of the invention, a DDA-based multi-touchpoint advertising performance feedback method includes:

[0059] Step S100: Obtain the click-through rate (CTR) of the ad, calculate the CTR decay rate, and calculate the fatigue coefficient based on the CTR decay rate, the close rate, and the number of times the ad has been exposed to the current user. The fatigue coefficient is an indicator used to quantify the degree of user boredom or discomfort with a specific ad ad. The value of this coefficient usually reflects the negative emotions of users caused by repeated exposure. The higher the value, the more fatigued the user is.

[0060] Step S200: Obtain the video completion rate, calculate the content coefficient based on the video completion rate, and calculate the tolerance coefficient based on the fatigue coefficient and the content coefficient. The content coefficient is an indicator used to evaluate the attractiveness or quality of the advertising creative itself. This coefficient is mainly based on the depth of user interaction with the advertising content, such as the completion rate of video ads. The higher the value, the stronger the attractiveness of the advertising content to users. The tolerance coefficient is an indicator used to measure the user's acceptance or tolerance for advertising display. This coefficient comprehensively considers the user's fatigue level and the attractiveness of the advertising content. The higher the value, the greater the user's tolerance for the advertisement and the more impressions they can accept.

[0061] Step S300: Obtain page dwell time and scroll depth, and calculate the value coefficient based on page dwell time and scroll depth; the value coefficient is an indicator used to evaluate the potential value or depth of interest shown by users during the current page browsing or interaction with advertisements. This coefficient is usually based on user behavior on the page, such as dwell time and scroll depth. The higher the value, the deeper the user interaction and the greater the potential value.

[0062] Step S400: Based on the preset base frequency, tolerance coefficient, and value coefficient, calculate the target display frequency upper limit and adjust the current display frequency upper limit to the target display frequency upper limit;

[0063] In the foregoing, the technical solution of this invention constructs a closed-loop, adaptive advertising frequency control architecture. It no longer relies on static rules or empirical thresholds, but instead dynamically calculates fatigue coefficients, content coefficients, tolerance coefficients, and value coefficients by acquiring multi-dimensional user behavior signals in real time (such as click-through rate decay rate, ignore / close rate, number of impressions, video completion rate, page dwell time, and scroll depth), and ultimately adjusts the upper limit of the target display frequency. This dynamic adjustment mechanism can continuously balance the advertiser's exposure needs with user experience, solving the problems of static control mechanisms, single feedback dimensions, and lack of scenario-based adaptation in existing technologies. It achieves the integration of advertising effectiveness evaluation and frequency control, thereby improving the overall efficiency of advertising and user acceptance.

[0064] Regarding step S100: The steps of obtaining the click-through rate (CTR) of the advertisement, calculating the CTR decay rate, and calculating the fatigue coefficient based on the CTR decay rate, the close rate, and the number of times the creative has been exposed to the current user include:

[0065] Get the click-through rate (CTR) of the ads and calculate the CTR decay rate;

[0066] Obtaining the close rate and the number of times the ad has been displayed to the current user; obtaining the click-through rate decay rate, ignore / close rate, and the number of times the ad has been displayed to the current user are key input parameters for assessing user fatigue with ad creatives; the click-through rate decay rate reflects the decreasing trend of user interest in a specific ad creative over time or with the number of exposures, and can be obtained by monitoring the changes in click-through rate at different exposure stages of the ad, for example, by comparing the current click-through rate with the historical average click-through rate or the baseline click-through rate. The ignore / close rate directly reflects the degree of negative feedback or disinterest from users towards the ad, and is usually statistically analyzed through user action data such as clicking "not interested" or "close ad". The number of times the ad has been displayed to the current user is a direct indicator of the frequency with which users encounter the ad creative, and can be accumulated and counted through the ad delivery system by tracking the exposure records of each user and each ad creative;

[0067] The current click-through rate (CTR) decay rate is normalized using a maximum-minimum method to obtain the CTR decay rate index. The normalization process is similar to that described above, transforming the original CTR decay rate data to a uniform, comparable scale, such as the range of 0 to 1. This normalization process eliminates differences in the dimensions and numerical ranges of CTR decay rates between different ad creatives or user groups, allowing this indicator to be effectively integrated with other fatigue indicators. For example, a linear normalization method can be used, subtracting the historical minimum value from the CTR decay rate and then dividing by the difference between the historical maximum and minimum values.

[0068] The current closure rate is compared with a preset threshold, and the ratio is limited to 1 using a min function to obtain the closure rate index. This process is to quantify the intensity of negative user feedback. By comparing it with the preset threshold, it can be determined whether the current ignore / closure rate is high or low relative to an acceptable level. Limiting the ratio to 1 using a min function can prevent extremely high ignore / closure rates from having an excessive impact on the fatigue coefficient, ensuring that the index is within a reasonable range, such as 0 to 1.

[0069] The creative exposure saturation is obtained by comparing the number of times the creative has been exposed to current users with the maximum number of times the creative's novelty has been maintained. This process measures the degree to which the "freshness" of the ad creative is depleted in the user's eyes. The maximum number of times the creative's novelty has been maintained is a preset, empirical threshold, representing the maximum number of times an ad creative can maintain a sense of novelty in the user's eyes. By comparing the numbers, a saturation value between 0 and 1 can be obtained. The higher the value, the lower the novelty of the creative for the user, and the more likely the user is to feel fatigued.

[0070] Based on the click-through rate decay rate index, the close rate index, and the creative exposure saturation, a fatigue coefficient is obtained by weighting and summing these three factors using preset weighting coefficients. The sum of each weighting coefficient is 1, and the fatigue coefficient ranges from 0 to 1, reflecting the degree of user fatigue with the advertisement. The specific calculation method is as follows: substitute the click-through rate decay rate index, the ignore / close rate index, and the creative exposure saturation into the formula. Obtain fatigue coefficient fatigue coefficient Output range 0-1, fatigue coefficient The larger the value, the more fatigued the user is, and the lower the user's interest in the advertisement. , and All are weighting coefficients with values ​​ranging from 0 to 1, and , This refers to the click-through rate decay rate index. To ignore / close rate index, This is for creative exposure saturation. The formula combines the three processed fatigue indicators mentioned above through a weighted summation, forming a unified fatigue coefficient. Among them, , and These are weighting coefficients, which allow the system to flexibly adjust the contribution of different fatigue indicators to the final fatigue coefficient based on actual business needs or experience. For example, these weights can be set based on historical data analysis of the importance of different indicators to user behavior. (Fatigue coefficient) The output range is 0-1. The larger the value, the higher the user's fatigue with the advertisement and the lower their interest.

[0071] The above solution is a multi-dimensional and refined method for calculating fatigue coefficients, which improves the accuracy of subsequent tolerance coefficient calculations and makes the adjustment of the upper limit of ad display frequency more scientific and reasonable. It not only effectively avoids user aversion and experience decline caused by excessive ad exposure, but also ensures that ads can be fully displayed while users still have interest, thereby optimizing the ad placement effect and achieving a win-win situation for user experience and ad revenue.

[0072] Specifically, regarding the click-through rate decay rate mentioned above, this parameter is calculated based on the click-through rate. The steps for obtaining the ad's click-through rate and calculating the click-through rate decay rate include:

[0073] Obtaining the current click-through rate (CTR) and baseline CTR: The current CTR refers to the ratio of actual clicks to impressions for a specific ad creative or user within a specific time window. This CTR reflects the user's recent interest in the ad creative. For example, it can be the CTR of the ad creative for the current user over the past 24 hours, or the CTR of the ad creative for the current user in the last N impressions. The baseline CTR is a reference standard used to measure the performance of an ad creative. It can be the historical average CTR of the ad creative among all users, the historical average CTR of the same user for similar ad creatives, or the average CTR of similar ads in the industry.

[0074] The current click-through rate (CTR) is compared to the baseline CTR, and the complement of this ratio is used as the CTR decay rate. The complement is one minus the ratio. By setting a stable baseline, the relative change in the current CTR can be better assessed. Rating the current CTR to the baseline CTR aims to quantify the deviation of the current CTR from the baseline. By calculating the ratio, it's easy to see whether the current ad performance is better than, equal to, or worse than the baseline. For example, a ratio greater than 1 indicates good performance; a ratio less than 1 indicates poor performance. Using the complement of the ratio as the CTR decay rate (one minus the ratio) transforms the relative change in CTR into an indicator of the degree of "decline" or "reduction." When the current CTR is lower than the baseline CTR, the ratio is less than 1, and its complement (1 minus the ratio) will be a positive value, indicating decay. When the current CTR is equal to or higher than the baseline CTR, the complement will be zero or negative, indicating no decay or even an increase. This approach allows the click-through rate decay rate to directly reflect the declining trend of user interest; the higher the value, the greater the user fatigue.

[0075] The click-through rate decay rate calculated above, by comparing the current click-through rate with the baseline click-through rate and quantifying the difference as a decay rate, can more accurately reflect changes in user interest and fatigue levels with ad creatives; this precise click-through rate decay rate serves as a fatigue coefficient. An important component that makes the fatigue coefficient The calculations are more reliable, thus improving the tolerance coefficient. The accuracy of the assessment; ultimately, based on a more accurate tolerance coefficient. The calculated upper limit of the target display frequency It allows for more precise adjustment of ad display frequency, effectively avoiding user aversion caused by excessive ad exposure or insufficient user reach due to insufficient exposure, thereby optimizing user experience and improving the overall effectiveness of ad placement.

[0076] Regarding step S200, the steps of obtaining the video completion rate, calculating the content coefficient based on the video completion rate, and calculating and obtaining the tolerance coefficient based on the fatigue coefficient and the content coefficient include:

[0077] Obtaining video completion rate is a key metric for measuring the completeness of a user's viewing of video content, reflecting their interest and engagement. This metric can be obtained by integrating a video player SDK into the user's device to monitor video playback progress and status in real time, and reporting playback completion events to the backend server for statistical analysis; alternatively, it can be obtained by analyzing user viewing log data and calculating the ratio of the user's actual viewing time to the total video duration.

[0078] Based on the video completion rate, it is converted into a content coefficient using a non-linear mapping function. This function centers on a benchmark completion rate threshold, and the slope parameter controls the steepness of the mapping curve. The output range is between 0 and 1, reflecting the attractiveness of the advertising content. The specific calculation method is as follows: substitute the video completion rate into the formula... Obtain content coefficient Content coefficient Output range 0-1, content coefficient The larger the value, the more attractive the content. It is a natural constant. The slope coefficient controls the steepness of the S-curve. For video completion rate, This serves as the baseline completion rate threshold. This step aims to adjust the original video completion rate. Convert to a content coefficient between 0 and 1 This is used to quantify the attractiveness of video content. The formula employed is an S-shaped curve function (Sigmoid function), characterized by its ability to simulate the non-linear trend of user interest gradually saturating from low to high. When the video completion rate... When the content coefficient is low, Slow growth; when Approaching the baseline completion rate threshold At that time, the content coefficient The most rapid growth indicates that changes in user interests have the greatest impact on content attractiveness assessment within this range; when When it is very high, the content coefficient Approaching saturation, growth slowed again. Among them, It is a natural constant and a fundamental component of this mathematical model; The slope coefficient is used to control the steepness of the S-curve; a larger value indicates a steeper slope. The value makes the curve at The changes in the vicinity were more dramatic, reflecting the sensitivity of content appeal to changes in completion rates; The baseline completion rate threshold represents the content coefficient. The completion rate of 0.5 can be set based on the average completion performance of different types of video ads. This calculation process can be implemented in the back-end processing module of the ad performance feedback system, using preset parameters. and It calculates the video completion rate data in real time or batch processing.

[0079] In the above, by employing an S-curve function to non-linearly transform video completion rates, a more accurate and refined capture of users' true interest and engagement with video content can be obtained, resulting in a more representative content coefficient. The accuracy of this content coefficient directly improves the reliability of the tolerance coefficient calculation, enabling the system to more intelligently balance user fatigue and content appeal when adjusting the target display frequency ceiling. Specifically, for high-quality, highly engaging video content, even if users have watched it multiple times, the system can appropriately increase the display frequency based on a high content coefficient without causing user aversion, thereby maximizing advertising value. Conversely, for content with lower appeal, the system will more strictly control the display frequency to avoid overexposure leading to user fatigue. This refined content appeal assessment mechanism significantly improves the intelligence level of ad frequency control and user experience, effectively avoiding inefficient ad delivery or user churn caused by inaccurate content assessment.

[0080] Based on the known fatigue coefficient and content coefficient, the fatigue coefficient F reflects the degree of user aversion to a specific advertising creative, while the content coefficient C measures the attractiveness of the advertising content to users. The fatigue coefficient F can be calculated or predicted using data analysis models based on historical user interaction data with the advertisement, such as click-through rate decay, ignore or close behaviors, and number of impressions. The content coefficient C can be evaluated based on metrics such as the advertisement's completion rate and interaction rate, or obtained through content analysis of the advertising creative itself. In another implementation, the fatigue coefficient F can be determined in real time using a pre-set rule engine, combined with user behavior patterns such as multiple exposures to the same advertisement within a short period without effective interaction. The content coefficient C can be comprehensively scored by analyzing characteristics such as the advertisement's video length, clarity, and thematic relevance, combined with historical user feedback data.

[0081] Substitute the fatigue coefficient and content coefficient into the formula. Obtain tolerance coefficient Tolerance coefficient Output range 0-1, tolerance coefficient A larger value indicates a higher tolerance level, and the frequency can be adjusted to maximize value. The fatigue coefficient, For content coefficients, This is a content compensation factor that controls the degree to which content quality compensates for user fatigue. The formula provides a mathematical model for quantifying user tolerance, comprehensively considering user fatigue levels. and content appeal It also introduced a content compensation factor. Adjustments can be made. The calculation of this formula can be performed by a dedicated calculation module on the backend server of the advertising delivery system, when the data is obtained... and After obtaining the value, directly substitute it into the formula for calculation. Alternatively, the calculation of this formula can be integrated into the real-time decision engine, and before each ad display, based on the latest... and Value dynamic calculation Tolerance coefficient The output range is limited to 0-1, which allows for unified quantization comparisons and calculations with other coefficients, ensuring its stability and interpretability throughout the frequency adjustment model. Tolerance coefficient The higher the value, the higher the user's tolerance for advertising, allowing the system to adjust ad display frequency based on the principle of maximizing value. Content Compensation Factor Used to control the degree to which content quality compensates for fatigue. It can be a preset constant, optimized and determined through offline experiments and A / B testing. In another implementation, It can be a dynamic parameter that is adjusted in real time through machine learning models based on factors such as ad type, user group, and historical data, in order to more finely control the degree of compensation.

[0082] In the above, by comprehensively considering both user fatigue with ads and the attractiveness of the ad content itself, and by introducing a content compensation factor to adjust content quality, the calculated tolerance coefficient T more accurately reflects users' true acceptance of ads. This avoids the one-sidedness or crudeness of traditional methods in assessing user tolerance, thus enabling a more intelligent balance between user experience and ad effectiveness in subsequent ad frequency adjustments. When user fatigue is high but content attractiveness is strong, the system can appropriately increase the tolerance based on the content compensation factor, avoiding missing opportunities to display high-quality content due to excessive frequency restrictions; conversely, when user fatigue is high and content attractiveness is low, the system can effectively decrease the tolerance, promptly reducing ad exposure, significantly improving user experience, and reducing user aversion to ads.

[0083] Regarding step S300, the step of obtaining the page dwell time and scroll depth, and calculating the value coefficient based on the page dwell time and scroll depth includes:

[0084] Acquiring page dwell time and scroll depth refers to the system collecting user behavior data on advertising pages in real-time or offline through a data acquisition module. Page dwell time refers to the total time a user spends on an advertising page from entering it to leaving, reflecting the user's initial interest and level of engagement with the page content. Scroll depth refers to the extent to which a user scrolls vertically across the page, usually expressed as a percentage of the total page height, reflecting the breadth and detail of the user's browsing of the page content. This data can be obtained by embedding front-end monitoring scripts (such as JavaScript) in the advertising page or through server log analysis.

[0085] The page dwell time is subjected to max-min normalization and then limited to a range of 0-1 to obtain the page dwell time index. This step transforms the raw page dwell time data into a standardized, dimensionless metric. Max-min normalization is a commonly used data preprocessing technique that linearly maps data to a specified range, such as [0,1]. Minimum and maximum are the historically observed minimum and maximum page dwell times, respectively. Limiting to 0-1 ensures that even in extreme cases (such as dwell time exceeding the historical maximum), the page dwell time index remains between 0 and 1, avoiding the impact of outliers on subsequent calculations.

[0086] The current scroll depth is processed against its maximum value to obtain a scroll depth index. This step transforms the raw scroll depth data into an indicator that intuitively reflects the user's browsing activity. For example, the current scroll depth can be directly divided by the total page height to obtain a percentage value between 0 and 1 as the scroll depth index. Alternatively, it can be segmented based on a preset scroll depth threshold, mapping different scroll depths to different index values. This approach makes the scroll depths of pages of different lengths comparable, thus more accurately measuring user interest in the content.

[0087] Based on the page dwell time index and scroll depth index, the two are weighted and combined using configurable weight parameters, and the result is normalized to obtain a value coefficient ranging from 0 to 1. This coefficient reflects the depth of user interaction on the current page. Specifically, the calculation method is as follows: substitute the page dwell time index and scroll depth index into the formula... Obtain value coefficient Value coefficient Output range 0-1, value coefficient The larger the value, the deeper the interaction. This is a page dwell time index. The rolling depth index, The value weight ranges from 0 to 1. This formula is a weighted average model used to comprehensively consider the impact of page dwell time and scroll depth on user value. The value weight... The system can adjust the relative importance of page dwell time and scroll depth in value coefficient calculation based on actual business needs or experience. For example, if the time a user spends on the page is considered more important than actual scrolling, the system can adjust the relative importance of these factors. Set it to a value greater than 0.5; conversely, if you believe that scroll depth better reflects the user's interest in the content, you can set it to a value greater than 0.5. Set it to a value less than 0.5. This allows for the flexible construction of a value coefficient that reflects the depth of user interaction. Its output range is 0-1, and the larger the value, the higher the user value and the deeper the interaction;

[0088] In the above examples, this invention can transform raw, heterogeneous user behavior data (page dwell time and scroll depth) into unified, standardized value coefficients. This enables advertising systems to more precisely identify users' genuine interest in advertising content and the depth of their interaction, thereby avoiding inaccurate ad targeting caused by coarsely assessing user value. By providing a quantifiable and adjustable value assessment, this method can more effectively guide subsequent adjustments to ad display frequency, ensuring that high-value users receive more appropriate exposure while reducing ineffective disturbance to low-value users, significantly improving ad delivery efficiency and user experience.

[0089] Regarding step S400: The step of calculating the target display frequency upper limit based on the preset base frequency, tolerance coefficient, and value coefficient, and adjusting the current display frequency upper limit to the target display frequency upper limit includes:

[0090] Obtain the base frequency, tolerance coefficient, and value coefficient. The base frequency can be understood as the default display frequency of an ad in the absence of user feedback or specific contexts; it is usually preset by the advertiser or platform based on ad type, delivery strategy, etc. The tolerance coefficient reflects the user's acceptance of the ad; a higher value indicates that the user is less likely to feel fatigued and has a greater tolerance for the ad. The value coefficient quantifies the potential value generated by user interaction with the ad; a higher value indicates deeper user interaction and higher ad value. Obtaining these coefficients is a prerequisite for accurately calculating the target display frequency ceiling, ensuring the comprehensiveness and accuracy of the calculation. For example, these coefficients can be obtained from real-time user behavior data analysis systems, historical ad performance databases, or preset configuration parameters.

[0091] Based on the base frequency, adjustments are made using a tolerance coefficient, and gain is further adjusted based on the value coefficient to obtain the upper limit of the target display frequency. The specific calculation method is as follows: substitute the base frequency, tolerance coefficient, and value coefficient into the formula. Get the target display frequency limit ,in, Based on frequency, Tolerance coefficient, This is the value gain coefficient. This is the value coefficient. This step is the core calculation in this solution. It uses a structured mathematical model to integrate multiple influencing factors to quantitatively determine the upper limit of the target display frequency for the advertisement. The formula is designed to balance user experience and advertising effectiveness, where the base frequency... Provided benchmarks and tolerance levels The base frequency is adjusted to reflect user acceptance of the advertisement, while the value coefficient... Then through the value gain coefficient Further increasing or decreasing the frequency to reward high-value interactions. This computational approach ensures that determining the upper limit of target display frequency is no longer based on empirical guesswork, but rather on data-driven quantitative decisions. For example, this calculation can be performed by a frequency control module within an ad delivery system, or by a separate decision engine. Base frequency is the initial or default frequency for ad delivery. It represents the number of times an ad is allowed to be shown without considering personalized user feedback. This parameter is typically preset based on the ad's business objectives, ad placement characteristics, or industry standards. The tolerance coefficient is a value between 0 and 1 used to measure a user's acceptance of a particular advertisement. The higher the coefficient, the lower the user's fatigue with the advertisement, or the higher their interest in the advertisement content, thus allowing the advertisement to be displayed more frequently. The value gain coefficient is a weighted parameter used to adjust the degree to which the value coefficient affects the upper limit of the target display frequency. This coefficient can be configured according to the advertiser's campaign strategy or the platform's emphasis on user interaction. The value coefficient, a value between 0 and 1, is used to quantify the potential value generated by user interaction with an advertisement. The higher the coefficient, the deeper and more meaningful the user interaction with the advertisement.

[0092] Through the above technical solution, the present invention provides a quantitative and controllable method for calculating the upper limit of advertising target display frequency. This method integrates basic frequency, user tolerance, and advertising interaction value with a structured mathematical formula, overcoming the limitations of traditional frequency control methods that struggle to effectively quantify and comprehensively consider multi-dimensional influencing factors. Specifically, this solution can dynamically adjust the upper limit of advertising display frequency based on user fatigue with the advertisement (reflected by the tolerance coefficient) and the depth of user interaction with the advertisement (reflected by the value coefficient). This allows the advertising delivery system to more accurately balance user experience and advertising effectiveness, avoiding user aversion due to overexposure, while fully utilizing users' acceptance of high-value advertisements to maximize the effective reach and conversion potential of advertisements without causing user discomfort.

[0093] Figure 2 This is a block diagram of the composition structure of a DDA-based multi-touchpoint advertising performance feedback system. In this embodiment of the invention, a DDA-based multi-touchpoint advertising performance feedback system 10 includes:

[0094] The fatigue level determination module is used to obtain the click-through rate of the advertisement, calculate the click-through rate decay rate, and calculate the fatigue coefficient based on the click-through rate decay rate, the close rate, and the number of times the creative has been exposed to the current user.

[0095] The tolerance assessment module is used to obtain the video completion rate, calculate the content coefficient based on the video completion rate, and calculate and obtain the tolerance coefficient based on the fatigue coefficient and the content coefficient.

[0096] The value assessment module is used to obtain page dwell time and scroll depth, and calculate the value coefficient based on page dwell time and scroll depth;

[0097] The ad frequency adjustment module is used to calculate the target display frequency limit based on preset base frequency, tolerance coefficient, and value coefficient, and adjust the current display frequency limit to the target display frequency limit.

[0098] 1. The DDA-based multi-touchpoint advertising effect feedback system according to claim 7, characterized in that the fatigue determination module comprises:

[0099] The click-through rate analysis unit is used to obtain the click-through rate of advertisements and calculate the click-through rate decay rate;

[0100] The parameter acquisition unit is used to obtain the close rate and the number of times the creative has been exposed to the current user;

[0101] The decay index processing unit is used to perform maximum-min normalization on the current click-through rate decay rate to obtain the click-through rate decay rate index.

[0102] The closing index processing unit is used to process the ratio of the current closing rate to a preset threshold, and then use the min function to limit the upper limit of the ratio to 1 to obtain the closing rate index.

[0103] The saturation output unit is used to process the ratio of the number of times the creative has been exposed to the current user to the maximum number of times the creative's freshness has been reached, and obtain the creative's exposure saturation.

[0104] The weighted summation unit is used to sum the click-through rate decay rate index, the close rate index, and the creative exposure saturation using preset weighting coefficients to obtain a fatigue coefficient. The sum of each weighting coefficient is 1, and the fatigue coefficient ranges from 0 to 1, reflecting the degree of user fatigue with the advertisement.

[0105] 2. The DDA-based multi-touchpoint advertising effect feedback system according to claim 7, characterized in that the tolerance assessment module includes:

[0106] The completion rate acquisition unit is used to obtain the video completion rate;

[0107] The content analysis unit is used to convert video completion rate into content coefficients through a non-linear mapping function. The non-linear mapping function is centered on the benchmark completion rate threshold, and the steepness of the mapping curve is controlled by the slope parameter. The output range is between 0 and 1, reflecting the attractiveness of the advertising content.

[0108] The tolerance coefficient generation unit is used to substitute the fatigue coefficient and content coefficient into the formula. Obtain tolerance coefficient Tolerance coefficient Output range 0-1, tolerance coefficient A larger value indicates a higher tolerance level, and the frequency can be adjusted to maximize value. The fatigue coefficient, For content coefficients, This is a content compensation factor that controls the degree to which content quality compensates for fatigue.

[0109] 3. The DDA-based multi-touchpoint advertising performance feedback system according to claim 7, characterized in that the value assessment module includes:

[0110] The page information acquisition unit is used to obtain the page dwell time and scroll depth;

[0111] The dwell time index acquisition unit is used to perform maximum-minimum normalization on the page dwell time, and after limiting the range to 0-1, obtain the page dwell time index.

[0112] The scrolling index determination unit is used to process the current scrolling depth with its maximum value to obtain the scrolling depth index;

[0113] The value coefficient generation unit is used to perform weighted comprehensive processing on the page dwell time index and scroll depth index through configurable weight parameters, and normalize the result to obtain a value coefficient between 0 and 1. This coefficient reflects the user's interaction depth on the current page.

[0114] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for providing multi-touchpoint advertising performance feedback based on DDA, characterized in that, The method includes: Get the click-through rate of the ad, calculate the click-through rate decay rate, and calculate the fatigue coefficient based on the click-through rate decay rate, the close rate, and the number of times the ad has been exposed to the current user; Obtain the video completion rate, calculate the content coefficient based on the video completion rate, and calculate the acquisition tolerance coefficient based on the fatigue coefficient and the content coefficient; Obtain the page dwell time and scroll depth, and calculate the value coefficient based on the page dwell time and scroll depth; Based on the preset base frequency, tolerance coefficient, and value coefficient, calculate the target display frequency limit and adjust the current display frequency limit to the target display frequency limit.

2. The DDA-based multi-touchpoint advertising effect feedback method according to claim 1, characterized in that, The steps of obtaining the click-through rate (CTR) of the advertisement, calculating the CTR decay rate, and calculating the fatigue coefficient based on the CTR decay rate, the close rate, and the number of times the ad has been exposed to the current user include: Get the click-through rate (CTR) of the ads and calculate the CTR decay rate; Get the close rate and the number of times the creative has been exposed to the current user; The current click-through rate decay rate is processed by maximum-min normalization to obtain the click-through rate decay rate index. The current closure rate is compared with a preset threshold, and the closure rate index is obtained by using the min function to limit the upper limit of the ratio to 1. The creative exposure saturation is obtained by comparing the number of times the creative has been exposed to the current user with the maximum number of times the creative's freshness has been reached. Based on the click-through rate decay rate index, the close rate index, and the creative exposure saturation, the three factors are weighted and summed using preset weighting coefficients to obtain the fatigue coefficient. The sum of each weighting coefficient is 1, and the fatigue coefficient ranges from 0 to 1, reflecting the degree of user fatigue with the advertisement.

3. The DDA-based multi-touchpoint advertising effect feedback method according to claim 2, characterized in that, The steps of obtaining the click-through rate (CTR) of the advertisement and calculating the CTR decay rate include: Get the current click-through rate and the baseline click-through rate; The current click-through rate is compared with the baseline click-through rate, and the complement of the ratio is taken as the click-through rate decay rate.

4. The DDA-based multi-touchpoint advertising effect feedback method according to claim 1, characterized in that, The steps of obtaining the video completion rate, calculating the content coefficient based on the video completion rate, and calculating the tolerance coefficient based on the fatigue coefficient and the content coefficient include: Get the video completion rate; Based on the video completion rate, it is converted into content coefficients through a non-linear mapping function. The non-linear mapping function is centered on the benchmark completion rate threshold, and the steepness of the mapping curve is controlled by the slope parameter. The output range is between 0 and 1, reflecting the attractiveness of the advertising content. Substitute the fatigue coefficient and content coefficient into the formula. Obtain tolerance coefficient Tolerance coefficient Output range 0-1, tolerance coefficient A larger value indicates a higher tolerance level, and the frequency can be adjusted to maximize value. The fatigue coefficient, For content coefficients, This is a content compensation factor that controls the degree to which content quality compensates for fatigue.

5. The DDA-based multi-touchpoint advertising effect feedback method according to claim 1, characterized in that, The steps of obtaining page dwell time and scroll depth, and calculating the value coefficient based on page dwell time and scroll depth include: Get the page dwell time and scroll depth; The page dwell time is normalized by the maximum and minimum, and then limited to 0-1 to obtain the page dwell time index. The current scroll depth is processed and its maximum value is compared to obtain the scroll depth index; Based on the page dwell time index and scroll depth index, the two are weighted and combined using configurable weight parameters, and the results are normalized to obtain a value coefficient ranging from 0 to 1. This coefficient reflects the depth of user interaction on the current page.

6. The DDA-based multi-touchpoint advertising effect feedback method according to claim 1, characterized in that, The steps of calculating the target display frequency ceiling based on preset base frequency, tolerance coefficient, and value coefficient, and adjusting the current display frequency ceiling to the target display frequency ceiling, include: Obtain the basic frequency, tolerance coefficient, and value coefficient; Based on the base frequency, adjustments are made using the tolerance coefficient, and the gain is adjusted based on the value coefficient to obtain the upper limit of the target display frequency.

7. A multi-touchpoint advertising performance feedback system based on DDA, characterized in that, The system includes: The fatigue level determination module is used to obtain the click-through rate of the advertisement, calculate the click-through rate decay rate, and calculate the fatigue coefficient based on the click-through rate decay rate, the close rate, and the number of times the creative has been exposed to the current user. The tolerance assessment module is used to obtain the video completion rate, calculate the content coefficient based on the video completion rate, and calculate and obtain the tolerance coefficient based on the fatigue coefficient and the content coefficient. The value assessment module is used to obtain page dwell time and scroll depth, and calculate the value coefficient based on page dwell time and scroll depth; The ad frequency adjustment module is used to calculate the target display frequency limit based on preset base frequency, tolerance coefficient, and value coefficient, and adjust the current display frequency limit to the target display frequency limit.

8. The DDA-based multi-touchpoint advertising effect feedback system according to claim 7, characterized in that, The fatigue determination module includes: The click-through rate analysis unit is used to obtain the click-through rate of advertisements and calculate the click-through rate decay rate; The parameter acquisition unit is used to obtain the close rate and the number of times the creative has been exposed to the current user; The decay index processing unit is used to perform maximum-min normalization on the current click-through rate decay rate to obtain the click-through rate decay rate index. The closing index processing unit is used to process the ratio of the current closing rate to a preset threshold, and then use the min function to limit the upper limit of the ratio to 1 to obtain the closing rate index. The saturation output unit is used to process the ratio of the number of times the creative has been exposed to the current user to the maximum number of times the creative's freshness has been reached, and obtain the creative's exposure saturation. The weighted summation unit is used to sum the click-through rate decay rate index, the close rate index, and the creative exposure saturation using preset weighting coefficients to obtain a fatigue coefficient. The sum of each weighting coefficient is 1, and the fatigue coefficient ranges from 0 to 1, reflecting the degree of user fatigue with the advertisement.

9. The DDA-based multi-touchpoint advertising effect feedback system according to claim 7, characterized in that, The tolerance assessment module includes: The completion rate acquisition unit is used to obtain the video completion rate; The content analysis unit is used to convert video completion rate into content coefficients through a non-linear mapping function. The non-linear mapping function is centered on the benchmark completion rate threshold, and the steepness of the mapping curve is controlled by the slope parameter. The output range is between 0 and 1, reflecting the attractiveness of the advertising content. The tolerance coefficient generation unit is used to substitute the fatigue coefficient and content coefficient into the formula. Obtain tolerance coefficient Tolerance coefficient Output range 0-1, tolerance coefficient A larger value indicates a higher tolerance level, and the frequency can be adjusted to maximize value. The fatigue coefficient, For content coefficients, This is a content compensation factor that controls the degree to which content quality compensates for fatigue.

10. The DDA-based multi-touchpoint advertising effect feedback system according to claim 7, characterized in that, The valuation module includes: The page information acquisition unit is used to obtain the page dwell time and scroll depth; The dwell time index acquisition unit is used to perform maximum-minimum normalization on the page dwell time, and after limiting the range to 0-1, obtain the page dwell time index. The scrolling index determination unit is used to process the current scrolling depth with its maximum value to obtain the scrolling depth index; The value coefficient generation unit is used to perform weighted comprehensive processing on the page dwell time index and scroll depth index through configurable weight parameters, and normalize the result to obtain a value coefficient between 0 and 1. This coefficient reflects the user's interaction depth on the current page.

Images