A digital marketing system and method based on AIGC

Abstract

The present invention relates to the field of digital marketing technology, and discloses a digital marketing system and method based on AIGC, the system comprising: an advertising unit, collecting historical data and advertising type data, determining the benchmark exposure duration of an advertisement, and automatically generating and publishing an advertisement based on the AIGC technology; a collection unit; a processing unit, configured to calculate the comprehensive similarity according to the advertising performance data and the historical data when determining to adjust the initial advertising exposure duration, compare the comprehensive similarity with the comprehensive similarity threshold, and determine the exposure duration influencing factor according to the comprehensive similarity; an adjustment unit, configured to adjust the initial advertising exposure duration according to the exposure duration influencing factor, and complete the advertising delivery with the adjusted final advertising exposure duration. The present invention improves the return rate of overall advertising delivery, can respond to market changes in real time, optimize advertising delivery strategies, and achieve more efficient advertising effects.

Description

Technical Field

[0001] The present invention relates to the field of digital marketing technology, and in particular to a digital marketing system and method based on AIGC. Background Art

[0002] In today's digital age, the efficiency and effectiveness of advertising are often affected and restricted by a variety of complex factors. Traditional advertising methods mainly rely on advertisers' experience and judgment and market research data. However, these methods are often not flexible and accurate enough when dealing with rapidly changing market environments and user behaviors. With the advancement of science and technology and the development of big data technology, advertising has gradually shifted from the traditional experience-driven model to the data-driven model in order to improve the accuracy and efficiency of advertising.

[0003] In the existing digital marketing system, the efficiency and effectiveness of advertising are often affected and restricted by a variety of factors. These factors include but are not limited to the rapid changes in the market environment, the diversification and personalized needs of user behavior, the technical capabilities of the advertising platform, and the creativity and attractiveness of the advertising content. Traditional advertising methods mainly rely on the advertiser's experience and judgment and the support of market research data, but these methods are often not flexible and accurate enough when dealing with the rapidly changing market environment and user behavior. Summary of the invention

[0004] In view of this, the present invention proposes a digital marketing system and method based on AIGC, aiming to solve the problem that the current advertising delivery methods mainly rely on the advertiser's experience and judgment and the support of market research data, but these methods are often not flexible and accurate enough when dealing with rapidly changing market environments and user behaviors.

[0005] In one aspect, the present invention proposes a digital marketing system based on AIGC, comprising:

[0006] Ad units collect historical data and ad type data, determine the benchmark ad exposure duration, and automatically generate and publish ads based on AIGC technology;

[0007] A collection unit is configured to collect advertisement performance data and determine initial advertisement exposure duration after a first time interval after advertisement release; and is also configured to collect user real-time behavior data and compare it with the historical data after a second time interval, and determine whether to adjust the initial advertisement exposure duration according to the comparison result;

[0008] a processing unit configured to, when determining to adjust the initial advertisement exposure duration, calculate a comprehensive similarity based on the advertisement performance data and the historical data, compare the comprehensive similarity with a comprehensive similarity threshold, and determine an exposure duration influencing factor based on the comprehensive similarity; when the comprehensive similarity is greater than the comprehensive similarity threshold, determine the exposure duration influencing factor based on the historical data; when the comprehensive similarity is less than or equal to the similarity threshold, determine the exposure duration influencing factor through a machine learning algorithm;

[0009] The adjustment unit is configured to adjust the initial advertisement exposure duration according to the exposure duration influencing factor, and complete the advertisement delivery with the adjusted final advertisement exposure duration.

[0010] Furthermore, the advertising performance data includes advertising click-through rate, advertising conversion rate and user interaction rate.

[0011] Furthermore, determining the benchmark exposure duration of an advertisement includes:

[0012] The user group characteristic data is collected and combined with the advertisement type data, and the differences in the responses of different user groups to different types of advertisements are analyzed through the AIGC technology, so as to determine the benchmark exposure duration of different types of advertisements.

[0013] Further, after the advertisement is released, when the advertisement performance data is collected after the first time interval and the initial advertisement exposure duration is determined, the initial advertisement exposure duration is obtained by the following formula:

[0014] Ti=Tb*(1+ω1*CTRt+ω2*CRt+ω3*Ent)

[0015] Among them, Ti represents the initial advertising exposure time, Tb represents the benchmark exposure time, CTRt represents the average click-through rate of this type of advertising in the first time interval, CRt represents the average conversion rate of this type of advertising in the first time interval, Ent represents the average user interaction rate of this type of advertising in the first time interval, ω1, ω2 and ω3 are weight coefficients, and the sum of ω1, ω2, ω3 is 1.

[0016] Furthermore, after the second time interval, the real-time behavior data of the user is collected and compared with the historical data, and whether to adjust the initial advertisement exposure duration is determined according to the comparison result, including:

[0017] When the historical data contains the same data as the real-time behavior data, the initial advertisement exposure duration is not adjusted;

[0018] When the historical data does not contain the same data as the real-time behavior data, the initial advertisement exposure duration is adjusted.

[0019] Further, when it is determined to adjust the initial advertisement exposure duration, calculating the comprehensive similarity based on the advertisement performance data and the historical data includes:

[0020] So=α1*sim(CTRc,CTRh)+α2*sim(CRc,CRh)+α3*sim(Enc,Enh)

[0021] Among them, So represents the comprehensive similarity, α1, α2 and α3 represent weight coefficients, and the sum of α1, α2, α3 is 1, CTRc represents the current advertisement click-through rate of this type of advertisement, CTRh represents the historical advertisement click-through rate of this type of advertisement, sim(CTRc, CTRh) represents the similarity between the current advertisement click-through rate and the historical advertisement click-through rate of this type of advertisement, CRc represents the current advertisement conversion rate of this type of advertisement, CRh represents the historical advertisement conversion rate of this type of advertisement, sim(CRc, CRh) represents the similarity between the current advertisement conversion rate and the historical advertisement conversion rate of this type of advertisement, Enc represents the current advertisement user interaction rate of this type of advertisement, Enh represents the historical advertisement user interaction rate of this type of advertisement, and sim(Enc, Enh) represents the similarity between the current advertisement user interaction rate and the historical advertisement user interaction rate of this type of advertisement.

[0022] Furthermore, when determining the exposure duration influencing factors based on the comprehensive similarity, it includes:

[0023] The comprehensive similarity is compared with a preset comprehensive similarity threshold, and an exposure duration influencing factor is determined according to the comparison result;

[0024] When data with the comprehensive similarity greater than the comprehensive similarity threshold exists in the historical data, determining the exposure duration influencing factor according to the historical data;

[0025] When there is no data in the historical data whose comprehensive similarity is greater than the comprehensive similarity threshold, the exposure duration influencing factor is determined according to a machine learning algorithm.

[0026] Furthermore, when determining the exposure duration influencing factor according to the historical data, it includes:

[0027] When the data in the historical data whose comprehensive similarity is greater than the comprehensive similarity threshold is unique, the historical exposure duration influencing factor corresponding to the data is used as the exposure duration influencing factor;

[0028] When the data in the historical data whose comprehensive similarity is greater than the comprehensive similarity threshold is not unique, the average of the historical exposure duration influencing factors corresponding to the respective data is used as the exposure duration influencing factor.

[0029] Furthermore, when determining the exposure duration influencing factor according to the machine learning algorithm, it includes:

[0030] Establishing a training data set based on the advertising performance data, user behavior data, and historical exposure duration influencing factors;

[0031] Using a deep learning model, the relationship between the initial advertisement exposure duration and the user behavior data is learned through the training data set to predict the final advertisement exposure duration;

[0032] Gradient descent algorithm is used to optimize model parameters;

[0033] After the training is completed, the advertisement performance data and user behavior data are input into the trained deep learning model to obtain the exposure duration influencing factor;

[0034] The initial advertisement exposure duration is adjusted according to the exposure duration influencing factor.

[0035] Compared with the prior art, the beneficial effect of the present invention is that: in the present invention, after the advertisement is released, the collection unit can quickly evaluate whether the initial exposure time of the advertisement is appropriate by real-time monitoring and analyzing the performance data of the advertisement. After the first time interval, the collection unit will collect key indicators such as the click-through rate and conversion rate of the advertisement to determine whether the initial exposure time needs to be adjusted; after the second time interval, the collection unit further collects the user's real-time behavior data and compares it with the historical data to determine whether the performance of the advertisement meets expectations; when the processing unit determines that the initial advertisement exposure time needs to be adjusted, it will calculate the comprehensive similarity between the advertisement performance data and the historical data; when the comprehensive similarity is greater than the comprehensive similarity threshold, it means that the performance of the current advertisement is similar to the historical data, and the processing unit will determine the exposure time influencing factor based on the historical data; when the comprehensive similarity is less than or equal to the similarity threshold, it means that there is a large difference between the performance of the current advertisement and the historical data, and the processing unit will analyze the data through a machine learning algorithm to determine the exposure time influencing factor; the adjustment unit adjusts the initial advertisement exposure time according to the exposure time influencing factor to optimize the advertising effect. The adjusted final ad exposure time will help improve the click-through rate and conversion rate of the ad, thereby improving the return on the overall advertising. Through this dynamic adjustment mechanism, the digital marketing system can respond to market changes in real time, optimize advertising strategies, and achieve more efficient advertising results.

[0036] In another aspect, the present invention further proposes a digital marketing method based on AIGC, comprising the following steps:

[0037] S100: Collect historical data and advertising type data, determine the benchmark exposure duration of the advertisement, and automatically generate and publish advertisements based on AIGC technology;

[0038] S200: After the advertisement is published, the advertisement performance data is collected after a first time interval and the initial advertisement exposure duration is determined; it is also configured to collect the user's real-time behavior data after a second time interval and compare it with the historical data, and determine whether to adjust the initial advertisement exposure duration according to the comparison result;

[0039] S300: when it is determined that the initial advertisement exposure duration is to be adjusted, a comprehensive similarity is calculated based on the advertisement performance data and the historical data, the comprehensive similarity is compared with a comprehensive similarity threshold, and an exposure duration influencing factor is determined based on the comprehensive similarity; when the comprehensive similarity is greater than the comprehensive similarity threshold, the exposure duration influencing factor is determined based on the historical data; when the comprehensive similarity is less than or equal to the similarity threshold, the exposure duration influencing factor is determined by a machine learning algorithm;

[0040] S400: adjusting the initial advertisement exposure duration according to the exposure duration influencing factor, and completing advertisement delivery with the adjusted final advertisement exposure duration.

[0041] It is understandable that the above-mentioned AIGC-based digital marketing system and method have the same beneficial effects and will not be elaborated here. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Various other advantages and benefits will become apparent to those of ordinary skill in the art by reading the detailed description of the preferred embodiments below. The accompanying drawings are only for the purpose of illustrating the preferred embodiments and are not to be considered as limiting the present invention. Moreover, the same reference symbols are used throughout the accompanying drawings to represent the same components. In the accompanying drawings:

[0043] Figure 1 A structural block diagram of a digital marketing system based on AIGC provided in an embodiment of the present invention;

[0044] Figure 2 A flowchart of a digital marketing method based on AIGC provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0045] Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments described herein. On the contrary, these embodiments are provided in order to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art. It should be noted that, in the absence of conflict, the embodiments of the present invention and the features described in the embodiments can be combined with each other. The present invention will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0046] See also Figure 1 As shown, in some embodiments of the present application, this embodiment provides a digital marketing system based on AIGC, including:

[0047] Ad units collect historical data and ad type data, determine the benchmark ad exposure duration, and automatically generate and publish ads based on AIGC technology;

[0048] A collection unit is configured to collect advertisement performance data and determine initial advertisement exposure duration after a first time interval after advertisement release; and is also configured to collect user real-time behavior data and compare it with the historical data after a second time interval, and determine whether to adjust the initial advertisement exposure duration according to the comparison result;

[0049] a processing unit configured to, when determining to adjust the initial advertisement exposure duration, calculate a comprehensive similarity based on the advertisement performance data and the historical data, compare the comprehensive similarity with a comprehensive similarity threshold, and determine an exposure duration influencing factor based on the comprehensive similarity; when the comprehensive similarity is greater than the comprehensive similarity threshold, determine the exposure duration influencing factor based on the historical data; when the comprehensive similarity is less than or equal to the similarity threshold, determine the exposure duration influencing factor through a machine learning algorithm;

[0050] The adjustment unit is configured to adjust the initial advertisement exposure duration according to the exposure duration influencing factor, and complete the advertisement delivery with the adjusted final advertisement exposure duration.

[0051] It can be seen that in some embodiments of the present invention, the advertising unit is used to collect historical data and advertising type data, determine the benchmark exposure time of the advertisement, and automatically generate and publish advertisements based on AIGC technology. The full name of AIGC technology is Artificial Intelligence Generated Content. This technology is a content generation technology based on artificial intelligence, which can automatically generate creative content based on input data and parameters. In the digital marketing system of the present invention, AIGC technology is applied to advertisement generation, making the advertising content more diversified and personalized, thereby improving the attractiveness and effectiveness of the advertisement.

[0052] It can be seen that in some embodiments of the present invention, after the advertisement is released, the collection unit can quickly evaluate whether the initial exposure time of the advertisement is appropriate by real-time monitoring and analyzing the performance data of the advertisement. After the first time interval, the collection unit will collect key indicators such as the click-through rate and conversion rate of the advertisement to determine whether the initial exposure time needs to be adjusted; after the second time interval, the collection unit further collects the user's real-time behavior data and compares it with the historical data to determine whether the performance of the advertisement meets expectations; when the processing unit determines that the initial advertisement exposure time needs to be adjusted, it will calculate the comprehensive similarity between the advertisement performance data and the historical data; when the comprehensive similarity is greater than the comprehensive similarity threshold, it means that the performance of the current advertisement is similar to the historical data, and the processing unit will determine the exposure time influencing factor based on the historical data; when the comprehensive similarity is less than or equal to the similarity threshold, it means that there is a large difference between the performance of the current advertisement and the historical data, and the processing unit will analyze the data through a machine learning algorithm to determine the exposure time influencing factor; the adjustment unit adjusts the initial advertisement exposure time according to the exposure time influencing factor to optimize the advertising effect. The adjusted final advertisement exposure time will help improve the click-through rate and conversion rate of the advertisement, thereby improving the return on the overall advertising delivery. Through this dynamic adjustment mechanism, the digital marketing system can respond to market changes in real time, optimize advertising strategies, and achieve more efficient advertising results.

[0053] It is understandable that historical data includes historical advertising performance data and historical behavioral data; historical data provides a valuable information foundation for digital marketing systems. These data reflect the performance of past advertising campaigns and help the system understand which advertising elements and strategies perform well in specific markets or user groups. By analyzing historical data, the system can identify successful advertising patterns and user preferences, and thus use this information when generating new advertisements to improve the relevance and attractiveness of advertisements.

[0054] Specifically, the advertising performance data includes advertising click-through rate, advertising conversion rate and user interaction rate.

[0055] It can be seen that in some embodiments of the present invention, advertising performance data is a key indicator for measuring advertising effectiveness. The click-through rate of an advertisement reflects the user's interest in the advertisement, and a high click-through rate usually means that the advertising content is attractive; the advertising conversion rate is directly related to the actual effect of the advertisement, that is, the proportion of users who take the expected action (such as purchasing a product or registering for a service) after seeing the advertisement; the user interaction rate reflects the degree of interaction between the user and the advertising content, including likes, comments, and sharing. By comprehensively analyzing these data, the system can more comprehensively evaluate the performance of the advertisement and adjust the exposure time accordingly to achieve the best advertising effect.

[0056] Specifically, the benchmark exposure duration of an ad includes:

[0057] The user group characteristic data is collected and combined with the advertisement type data, and the differences in the responses of different user groups to different types of advertisements are analyzed through the AIGC technology, so as to determine the benchmark exposure duration of different types of advertisements.

[0058] It can be seen that in some embodiments of the present invention, in the process of determining the benchmark exposure time of an advertisement, the system will comprehensively consider the user group characteristic data and the advertisement type data. Through AIGC technology, the system can analyze the differences in the responses of different user groups to different types of advertisements, thereby setting a reasonable benchmark exposure time for each type of advertisement. This personalized setting based on user behavior and preferences helps ensure that the advertising content can more effectively attract the target audience and improve the overall effectiveness of the advertisement.

[0059] Specifically, after an advertisement is released, when the advertisement performance data is collected after a first time interval and the initial advertisement exposure duration is determined, the initial advertisement exposure duration is obtained by the following formula:

[0060] Ti=Tb*(1+ω1*CTRt+ω2*CRt+ω3*Ent)

[0061] Among them, Ti represents the initial advertising exposure time, Tb represents the benchmark exposure time, CTRt represents the average click-through rate of this type of advertising in the first time interval, CRt represents the average conversion rate of this type of advertising in the first time interval, Ent represents the average user interaction rate of this type of advertising in the first time interval, ω1, ω2 and ω3 are weight coefficients, and the sum of ω1, ω2, ω3 is 1.

[0062] It can be seen that in some embodiments of the present invention, after the first time interval, the system will collect advertising performance data, including key indicators such as click-through rate, conversion rate and user interaction rate. These data will be used to determine the initial advertising exposure time. In the calculation formula of the initial advertising exposure time, the weight coefficients ω1, ω2 and ω3 correspond to the click-through rate, conversion rate and user interaction rate respectively, and their sum is 1, which ensures that the impact of different indicators on the exposure time can be properly balanced. In this way, the system can dynamically adjust the exposure time according to the performance of the advertisement in actual delivery to achieve the best advertising effect. After the second time interval, the system will further collect the user's real-time behavior data and compare it with the historical data. This comparison helps the system determine whether the performance of the advertisement meets expectations and whether the initial exposure time needs to be adjusted. If the performance of the advertisement is similar to the historical data, the system will determine the exposure time influencing factor based on the historical data; if the performance of the advertisement is significantly different from the historical data, the system will analyze the data through a machine learning algorithm to determine the exposure time influencing factor. This data-driven dynamic adjustment mechanism enables the digital marketing system to respond to market changes in real time, optimize advertising delivery strategies, and achieve more efficient advertising effects.

[0063] It is understandable that in some embodiments of the present invention, the advertising content generated by AIGC technology is more diversified and personalized, and combined with the mechanism of real-time monitoring and analysis of advertising performance data, it can effectively improve the attractiveness and effectiveness of advertising. At the same time, by dynamically adjusting the exposure time, the system can ensure the efficiency and return rate of advertising delivery, and provide advertisers with more accurate and efficient digital marketing solutions.

[0064] Specifically, after the second time interval, the real-time behavior data of the user is collected and compared with the historical data, and whether to adjust the initial advertisement exposure duration is determined according to the comparison result, including:

[0065] When the historical data contains the same data as the real-time behavior data, the initial advertisement exposure duration is not adjusted;

[0066] When the historical data does not contain the same data as the real-time behavior data, the initial advertisement exposure duration is adjusted.

[0067] It can be seen that in some embodiments of the present invention, the system will make a critical judgment after the second time interval. If there is the same data in the historical data as in the real-time behavior data, it means that the performance of the advertisement is consistent with the historical trend, so there is no need to adjust the initial advertisement exposure duration. In this case, the system will maintain the current exposure strategy to ensure the stability and continuity of the advertisement delivery. However, if there is no data in the historical data that is the same as in the real-time behavior data, this indicates that there is a new trend or change in the performance of the advertisement. In this case, the system will adjust the initial advertisement exposure duration to adapt to the new market environment and user behavior. The adjustment process will be based on the difference between the real-time behavior data and the historical data, and the data will be analyzed by a machine learning algorithm to determine the new exposure duration influencing factor. This helps to ensure that the advertising delivery strategy can adapt to market changes in a timely manner and improve the advertising effect. Through this dynamic adjustment mechanism, the digital marketing system can respond to market changes in real time, optimize the advertising delivery strategy, and achieve more efficient advertising effects. Advertisers will be able to obtain higher returns, and users will also be exposed to more relevant and attractive advertising content. This data-driven advertising delivery mechanism not only improves advertising effects, but also brings new development opportunities to the field of digital marketing.

[0068] Specifically, when it is determined to adjust the initial advertisement exposure duration, the comprehensive similarity is calculated based on the advertisement performance data and the historical data, including:

[0069] So=α1*sim(CTRc,CTRh)+α2*sim(CRc,CRh)+α3*sim(Enc,Enh)

[0070] Among them, So represents the comprehensive similarity, α1, α2 and α3 represent weight coefficients, and the sum of α1, α2, α3 is 1, CTRc represents the current advertisement click-through rate of this type of advertisement, CTRh represents the historical advertisement click-through rate of this type of advertisement, sim(CTRc, CTRh) represents the similarity between the current advertisement click-through rate and the historical advertisement click-through rate of this type of advertisement, CRc represents the current advertisement conversion rate of this type of advertisement, CRh represents the historical advertisement conversion rate of this type of advertisement, sim(CRc, CRh) represents the similarity between the current advertisement conversion rate and the historical advertisement conversion rate of this type of advertisement, Enc represents the current advertisement user interaction rate of this type of advertisement, Enh represents the historical advertisement user interaction rate of this type of advertisement, and sim(Enc, Enh) represents the similarity between the current advertisement user interaction rate and the historical advertisement user interaction rate of this type of advertisement.

[0071] It can be seen that in some embodiments of the present invention, the system will further refine the analysis of the advertising performance data and determine whether the initial advertising exposure time needs to be adjusted by calculating the comprehensive similarity. In the calculation formula of the comprehensive similarity, the weight coefficients α1, α2 and α3 correspond to the similarity between the current advertising click-through rate, conversion rate and user interaction rate and the historical data respectively. The sum of these weight coefficients is 1, which ensures that the impact of different indicators on the similarity can be properly balanced. When calculating the comprehensive similarity, the system will calculate the similarity between the current advertising click-through rate and the historical advertising click-through rate (sim(CTRc, CTRh)), the similarity between the current advertising conversion rate and the historical advertising conversion rate (sim(CRc, CRh)), and the similarity between the current advertising user interaction rate and the historical advertising user interaction rate (sim(Enc, Enh)). In this way, the system can comprehensively consider the performance of the advertisement in different aspects, so as to more accurately evaluate whether the performance of the advertisement is consistent with the historical trend.

[0072] It is understandable that in some embodiments of the present invention, if the comprehensive similarity is high, indicating that the performance of the current advertisement is similar to the historical trend, the system will keep the initial exposure duration unchanged. On the contrary, if the comprehensive similarity is low, indicating that there is a large difference between the performance of the current advertisement and the historical trend, the system will adjust the initial exposure duration. The adjustment process will be based on the difference between the current advertisement performance data and the historical data, and the data will be analyzed by a machine learning algorithm to determine the new exposure duration influencing factor. This adjustment mechanism based on comprehensive similarity enables the digital marketing system to respond to market changes more finely and optimize the advertising delivery strategy. By dynamically adjusting the exposure duration, the system can ensure the efficiency and return rate of advertising delivery, and provide advertisers with more accurate and efficient digital marketing solutions. Advertisers will be able to obtain higher returns, and users will also be exposed to more relevant and attractive advertising content. This data-driven advertising delivery mechanism not only improves advertising effectiveness, but also brings new development opportunities to the field of digital marketing.

[0073] Specifically, when determining the exposure duration influencing factors based on comprehensive similarity, it includes:

[0074] The comprehensive similarity is compared with a preset comprehensive similarity threshold, and an exposure duration influencing factor is determined according to the comparison result;

[0075] When data with the comprehensive similarity greater than the comprehensive similarity threshold exists in the historical data, determining the exposure duration influencing factor according to the historical data;

[0076] When there is no data in the historical data whose comprehensive similarity is greater than the comprehensive similarity threshold, the exposure duration influencing factor is determined according to a machine learning algorithm.

[0077] It can be seen that in some embodiments of the present invention, the system will further refine the decision-making process for adjusting the exposure duration of the advertisement. By comparing the comprehensive similarity with a preset threshold, the system can determine whether the similarity between the current advertisement performance and the historical trend is high enough to determine whether the exposure duration needs to be adjusted. If there is data in the historical data with a comprehensive similarity greater than the comprehensive similarity threshold, this means that the performance of the current advertisement is relatively consistent with the historical trend, and the system will determine the exposure duration influencing factor based on the historical data. This step ensures that when the advertisement performance is relatively stable, the system can use historical experience to optimize the exposure duration, thereby maintaining the continuity and stability of advertisement delivery. However, if there is no data in the historical data with a comprehensive similarity greater than the comprehensive similarity threshold, this indicates that there is a large difference between the performance of the current advertisement and the historical trend, and the system will use a machine learning algorithm to determine the exposure duration influencing factor. In this way, the system can dynamically adjust the exposure duration to adapt to new trends according to changes in the current market environment and user behavior. The machine learning algorithm can extract valuable information from a large amount of data, identify the key factors affecting advertisement performance, and formulate a more accurate exposure duration strategy based on this.

[0078] It is understandable that in some embodiments of the present invention, this decision-making mechanism based on the comprehensive similarity threshold enables the digital marketing system to flexibly respond to market changes while maintaining the stability of advertising delivery. By dynamically adjusting the exposure duration, the system can ensure the efficiency and return rate of advertising delivery, and provide advertisers with more accurate and efficient digital marketing solutions. Advertisers will be able to obtain higher returns, and users will be exposed to more relevant and attractive advertising content. This data-driven advertising delivery mechanism not only improves advertising effectiveness, but also brings new development opportunities to the field of digital marketing.

[0079] Specifically, determining the exposure duration influencing factor according to the historical data includes:

[0080] When the data in the historical data whose comprehensive similarity is greater than the comprehensive similarity threshold is unique, the historical exposure duration influencing factor corresponding to the data is used as the exposure duration influencing factor;

[0081] When the data in the historical data whose comprehensive similarity is greater than the comprehensive similarity threshold is not unique, the average of the historical exposure duration influencing factors corresponding to the respective data is used as the exposure duration influencing factor.

[0082] It can be seen that in some embodiments of the present invention, the system will further refine the analysis of historical data to ensure the accuracy of exposure duration adjustment. When there are multiple data with comprehensive similarities greater than the comprehensive similarity threshold in the historical data, the system will calculate the mean of the historical exposure duration influencing factors corresponding to these data. This method can balance the impact of different historical data on the current exposure duration and avoid the contingency of a single data point from having too much impact on the decision. By taking the average, the system can comprehensively consider multiple historical trends to formulate a more robust exposure duration strategy. In addition, when there is only one data with a comprehensive similarity greater than the comprehensive similarity threshold in the historical data, the system will directly use the historical exposure duration influencing factor corresponding to the data. This method ensures that when the historical data is relatively consistent, the system can quickly and accurately determine the exposure duration influencing factor, thereby improving decision-making efficiency. In actual applications, the system will further optimize the calculation method of the exposure duration influencing factor through a machine learning algorithm based on the performance data and historical data of the current advertisement. For example, the system may introduce a time decay factor so that the recent data has a greater impact on the exposure duration, thereby better adapting to market changes. In addition, the system may also consider external factors such as seasonality and holiday effects to ensure the comprehensiveness and accuracy of the exposure duration adjustment strategy.

[0083] It is understandable that in some embodiments of the present invention, by comprehensively considering the similarity between current advertising performance data and historical data, combined with optimization methods such as machine learning algorithms and time decay factors, the system can dynamically adjust the advertising exposure duration, thereby achieving more accurate and efficient digital marketing. This data-driven advertising delivery mechanism not only improves advertising effectiveness, but also brings new development opportunities to the field of digital marketing. Advertisers will be able to obtain higher returns, and users will also be exposed to more relevant and attractive advertising content.

[0084] Specifically, when determining the exposure duration influencing factor according to the machine learning algorithm, it includes:

[0085] Establishing a training data set based on the advertising performance data, user behavior data, and historical exposure duration influencing factors;

[0086] Using a deep learning model, the relationship between the initial advertisement exposure duration and the user behavior data is learned through the training data set to predict the final advertisement exposure duration;

[0087] Gradient descent algorithm is used to optimize model parameters;

[0088] After the training is completed, the advertisement performance data and user behavior data are input into the trained deep learning model to obtain the exposure duration influencing factor;

[0089] The initial advertisement exposure duration is adjusted according to the exposure duration influencing factor.

[0090] It can be seen that in some embodiments of the present invention, the system will further use advanced machine learning technology to optimize the decision-making process of advertising exposure duration. By establishing a training data set containing advertising performance data, user behavior data, and historical exposure duration influencing factors, the system can train a deep learning model to learn the complex relationship between the initial advertising exposure duration and the user behavior data. This method enables the system to capture the subtle correlation between advertising performance and user behavior, thereby predicting the final advertising exposure duration. During the training process, the gradient descent algorithm is used to optimize the model parameters to ensure that the model can accurately reflect the relationship between the data. Through repeated iterations, the model gradually approaches the optimal solution, thereby improving the accuracy of the prediction. After the training is completed, the system inputs the latest advertising performance data and user behavior data into the trained deep learning model to obtain the exposure duration influencing factors. These influencing factors will be used to adjust the initial advertising exposure duration to achieve the best advertising effect.

[0091] It is understandable that without using historical data, after adjusting the initial ad exposure time through the algorithm, the ad performance data, user behavior data, exposure time influencing factors and adjusted exposure time are finally stored in the historical data, which will help the system to better use historical experience for optimization in future ad delivery. By continuously accumulating and analyzing this data, the system can continuously improve the ad delivery strategy and improve the ad effect and return on investment.

[0092] It is understandable that in some embodiments of the present invention, this exposure duration optimization method based on deep learning can not only cope with complex and changing market environments, but also dynamically adjust advertising exposure strategies according to real-time changes in user behavior. In this way, advertisers can ensure that their advertising resources are used most effectively, thereby maximizing the return on advertising investment. At the same time, users will also get a more personalized and high-quality advertising experience because the system can adjust the advertising content and exposure duration according to the user's actual interests and behavioral habits.

[0093] See also Figure 2 As shown, in some embodiments of the present application, this embodiment provides a digital marketing method based on AIGC, including the following steps:

[0094] S100: Collect historical data and advertising type data, determine the benchmark exposure duration of the advertisement, and automatically generate and publish advertisements based on AIGC technology;

[0095] S200: After the advertisement is published, the advertisement performance data is collected after a first time interval and the initial advertisement exposure duration is determined; it is also configured to collect the user's real-time behavior data after a second time interval and compare it with the historical data, and determine whether to adjust the initial advertisement exposure duration according to the comparison result;

[0096] S300: when it is determined that the initial advertisement exposure duration is to be adjusted, a comprehensive similarity is calculated based on the advertisement performance data and the historical data, the comprehensive similarity is compared with a comprehensive similarity threshold, and an exposure duration influencing factor is determined based on the comprehensive similarity; when the comprehensive similarity is greater than the comprehensive similarity threshold, the exposure duration influencing factor is determined based on the historical data; when the comprehensive similarity is less than or equal to the similarity threshold, the exposure duration influencing factor is determined by a machine learning algorithm;

[0097] S400: adjusting the initial advertisement exposure duration according to the exposure duration influencing factor, and completing advertisement delivery with the adjusted final advertisement exposure duration.

[0098] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems or computer program products. Therefore, the present application may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present application may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.

[0099] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems) and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0100] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0101] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0102] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the relevant field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A digital marketing system based on AIGC, characterized in that: include: Ad units collect historical data and ad type data, determine the benchmark ad exposure duration, and automatically generate and publish ads based on AIGC technology; A collection unit is configured to collect advertisement performance data and determine initial advertisement exposure duration after a first time interval after advertisement release; and is further configured to collect user real-time behavior data and compare the data with the historical data after a second time interval, and determine whether to adjust the initial advertisement exposure duration according to the comparison result; the advertisement performance data includes advertisement click-through rate, advertisement conversion rate and user interaction rate; a processing unit configured to, when determining to adjust the initial advertisement exposure duration, calculate a comprehensive similarity based on the advertisement performance data and the historical data, compare the comprehensive similarity with a comprehensive similarity threshold, and determine an exposure duration influencing factor based on the comprehensive similarity; When the comprehensive similarity is greater than the comprehensive similarity threshold, determining the exposure duration influencing factor according to the historical data; When the comprehensive similarity is less than or equal to the similarity threshold, determining the exposure duration influencing factor by a machine learning algorithm; When it is determined that the initial advertisement exposure duration is to be adjusted, calculating the comprehensive similarity based on the advertisement performance data and the historical data includes: So=α1*sim(CTRc,CTRh)+α2*sim(CRc,CRh)+α3*sim(Enc,Enh); Wherein, So represents the comprehensive similarity, α1, α2 and α3 represent weight coefficients, and the sum of α1, α2 and α3 is 1, CTRc represents the current advertisement click-through rate of this type of advertisement, CTRh represents the historical advertisement click-through rate of this type of advertisement, sim(CTRc, CTRh) represents the similarity between the current advertisement click-through rate and the historical advertisement click-through rate of this type of advertisement, CRc represents the current advertisement conversion rate of this type of advertisement, CRh represents the historical advertisement conversion rate of this type of advertisement, sim(CRc, CRh) represents the similarity between the current advertisement conversion rate and the historical advertisement conversion rate of this type of advertisement, Enc represents the current advertisement user interaction rate of this type of advertisement, Enh represents the historical advertisement user interaction rate of this type of advertisement, and sim(Enc, Enh) represents the similarity between the current advertisement user interaction rate and the historical advertisement user interaction rate of this type of advertisement; An adjustment unit, configured to adjust the initial advertisement exposure duration according to the exposure duration influencing factor, and complete advertisement delivery with the adjusted final advertisement exposure duration; When determining the benchmark exposure duration of an ad, include: The user group characteristic data is collected and combined with the advertisement type data, and the differences in the responses of different user groups to different types of advertisements are analyzed through the AIGC technology, so as to determine the benchmark exposure duration of different types of advertisements.

2. The AIGC-based digital marketing system according to claim 1, characterized in that: After the advertisement is released, when the advertisement performance data is collected after the first time interval and the initial advertisement exposure duration is determined, the initial advertisement exposure duration is obtained by the following formula: Ti=Tb*(1+ω1*CTRt+ω2*CRt+ω3*Ent) Among them, Ti represents the initial advertising exposure time, Tb represents the benchmark exposure time, CTRt represents the average click-through rate of this type of advertising in the first time interval, CRt represents the average conversion rate of this type of advertising in the first time interval, Ent represents the average user interaction rate of this type of advertising in the first time interval, ω1, ω2 and ω3 are weight coefficients, and the sum of ω1, ω2, ω3 is 1.

3. The AIGC-based digital marketing system according to claim 1, characterized in that: When determining the exposure duration influencing factors based on comprehensive similarity, it includes: The comprehensive similarity is compared with a preset comprehensive similarity threshold, and an exposure duration influencing factor is determined according to the comparison result; When data with the comprehensive similarity greater than the comprehensive similarity threshold exists in the historical data, determining the exposure duration influencing factor according to the historical data; When there is no data in the historical data whose comprehensive similarity is greater than the comprehensive similarity threshold, the exposure duration influencing factor is determined according to a machine learning algorithm.

4. The AIGC-based digital marketing system according to claim 3, characterized in that: When determining the exposure duration influencing factor according to the historical data, it includes: When the data in the historical data whose comprehensive similarity is greater than the comprehensive similarity threshold is unique, the historical exposure duration influencing factor corresponding to the data is used as the exposure duration influencing factor; When the data in the historical data whose comprehensive similarity is greater than the comprehensive similarity threshold is not unique, the average of the historical exposure duration influencing factors corresponding to the respective data is used as the exposure duration influencing factor.

5. The AIGC-based digital marketing system according to claim 4, characterized in that: When determining the exposure duration influencing factor according to the machine learning algorithm, it includes: Establishing a training data set based on the advertising performance data, user behavior data, and historical exposure duration influencing factors; Using a deep learning model, the relationship between the initial advertisement exposure duration and the user behavior data is learned through the training data set to predict the final advertisement exposure duration; Gradient descent algorithm is used to optimize model parameters; After the training is completed, the advertisement performance data and user behavior data are input into the trained deep learning model to obtain the exposure duration influencing factor; The initial advertisement exposure duration is adjusted according to the exposure duration influencing factor.

6. A digital marketing method based on AIGC, applied to the digital marketing system based on AIGC as claimed in any one of claims 1 to 5, characterized in that: include: Collect historical data and advertising type data, determine the benchmark exposure duration of advertisements, and automatically generate and publish advertisements based on AIGC technology; After the advertisement is published, the advertisement performance data is collected after a first time interval and the initial advertisement exposure duration is determined; it is also configured to collect the user's real-time behavior data after a second time interval and compare it with the historical data, and determine whether to adjust the initial advertisement exposure duration according to the comparison result; the advertisement performance data includes advertisement click-through rate, advertisement conversion rate and user interaction rate; When it is determined that the initial advertisement exposure duration is to be adjusted, a comprehensive similarity is calculated based on the advertisement performance data and the historical data, the comprehensive similarity is compared with a comprehensive similarity threshold, and an exposure duration influencing factor is determined based on the comprehensive similarity; When the comprehensive similarity is greater than the comprehensive similarity threshold, determining the exposure duration influencing factor according to the historical data; When the comprehensive similarity is less than or equal to the similarity threshold, determining the exposure duration influencing factor by a machine learning algorithm; When it is determined that the initial advertisement exposure duration is to be adjusted, calculating the comprehensive similarity based on the advertisement performance data and the historical data includes: So=α1*sim(CTRc,CTRh)+α2*sim(CRc,CRh)+α3*sim(Enc,Enh); Wherein, So represents the comprehensive similarity, α1, α2 and α3 represent weight coefficients, and the sum of α1, α2 and α3 is 1, CTRc represents the current advertisement click-through rate of this type of advertisement, CTRh represents the historical advertisement click-through rate of this type of advertisement, sim(CTRc, CTRh) represents the similarity between the current advertisement click-through rate and the historical advertisement click-through rate of this type of advertisement, CRc represents the current advertisement conversion rate of this type of advertisement, CRh represents the historical advertisement conversion rate of this type of advertisement, sim(CRc, CRh) represents the similarity between the current advertisement conversion rate and the historical advertisement conversion rate of this type of advertisement, Enc represents the current advertisement user interaction rate of this type of advertisement, Enh represents the historical advertisement user interaction rate of this type of advertisement, and sim(Enc, Enh) represents the similarity between the current advertisement user interaction rate and the historical advertisement user interaction rate of this type of advertisement; Adjusting the initial advertisement exposure duration according to the exposure duration influencing factor, and completing advertisement delivery with the adjusted final advertisement exposure duration; When determining the benchmark exposure duration of an ad, include: The user group characteristic data is collected and combined with the advertisement type data, and the differences in the responses of different user groups to different types of advertisements are analyzed through the AIGC technology, so as to determine the benchmark exposure duration of different types of advertisements.

Images