A multimedia advertising design platform based on the Internet of Things

Abstract

The invention discloses a multimedia advertisement design platform based on the Internet of Things, comprising a data collection module, an advertisement delivery module and an advertisement design module. The data collection module is used to collect comprehensive browsing data of users, feedback data of users and purchase records, and input historical advertisement comprehensive benefit data into the system. The advertisement delivery module is used to analyze the user's preference for browsing information on a web page, analyze the user's consumption willingness and consumption level, analyze the characteristics of the advertisement content entrusted by the advertiser, match the user with the advertisement according to the analysis result, analyze the browsing level preferred by the user, formulate an advertisement delivery plan according to the analysis result, further analyze the revenue effect of the delivery plan, and determine the advertisement delivery plan according to the revenue effect. The advertisement design module is used to communicate with the advertiser to modify the advertisement design, render the advertisement effect and display it to the advertiser. The invention has the characteristics of improving accuracy and advertisement benefit.

Description

Technical Field

[0001] The present invention relates to the technical field of advertisement design platforms, and in particular to a multimedia advertisement design platform based on the Internet of Things. Background Art

[0002] With the rapid development of Internet technology and the advancement of social media, traditional media can no longer meet the needs of advertising promotion. Therefore, existing technologies have begun to promote advertising through Internet advertising platforms.

[0003] At present, most advertising companies generally use Internet technology to formulate advertising delivery plans for advertisers for advertising delivery. However, due to the diversity of web user preferences and advertising content, existing technologies cannot accurately match user preferences with advertising content, resulting in poor accuracy of advertising delivery and low returns. Moreover, since users' consumption willingness is affected by their consumption level, existing technologies generally use periodic quantitative advertising delivery, which can easily result in a large number of advertisements being delivered when users have no consumption willingness, resulting in poor advertising effectiveness. Therefore, it is very necessary to design a multimedia advertising design platform based on the Internet of Things to improve accuracy and advertising effectiveness. Summary of the invention

[0004] The purpose of the present invention is to provide a multimedia advertising design platform based on the Internet of Things to solve the problems raised in the above background technology.

[0005] In order to solve the above technical problems, the present invention provides the following technical solutions: a multimedia advertising design platform based on the Internet of Things, comprising a data collection module, an advertising delivery module and an advertising design module, characterized in that: the data collection module is used to collect users' comprehensive browsing data, users' feedback data and purchase records, and enter historical advertising comprehensive benefit data into the system; the advertising delivery module is used to analyze users' preferences for browsing information on web pages, analyze users' consumption willingness and consumption level, analyze the characteristics of advertising content entrusted by advertisers, match users with advertisements according to the analysis results, analyze users' preferred browsing levels, formulate advertising delivery plans according to the analysis results and matching results, further analyze the revenue effect of the delivery plans, and determine the advertising delivery plans according to the revenue effect; the advertising design module is used to render advertising effects and display them to advertisers, communicate with advertisers through the communication module to modify advertising design, and the data collection module, the advertising delivery module and the advertising design module are connected to each other for communication;

[0006] The bias analysis module includes a user preference analysis submodule and an advertisement feature analysis submodule. The user preference analysis submodule is used to analyze the browsing data, consumption willingness and consumption level of users on the web page. The advertisement feature analysis submodule is used to analyze the advertisement content commissioned by the advertiser, and obtain the user preference features and advertisement features according to the analysis results.

[0007] The delivery analysis module includes a delivery level analysis submodule, a delivery plan formulation submodule and a benefit analysis submodule. The delivery level analysis submodule is used to analyze the browsing data and corresponding browsing time of users at each web page level, and obtain the browsing characteristics of users based on the analysis results. The delivery plan formulation submodule is used to analyze the budget of advertisers for advertising, allocate the budget and formulate an advertising delivery plan. The benefit analysis submodule is used to analyze historical advertising benefits, train prediction models, analyze the matching degree between users and advertisements, market changes and strategy changes, and revise the advertising benefit prediction results based on the analysis results.

[0008] According to the above technical solution, the data collection module includes a data collection module and a historical data entry module. The data collection module is used to collect users' comprehensive browsing data, users' feedback data and purchase records, and the historical data entry module is used to enter historical advertising comprehensive benefit data into the system.

[0009] According to the above technical solution, the advertising delivery module includes a bias analysis module, which is used to analyze the user's browsing data, retrieval data, consumption intention and consumption level on the web page, analyze the advertising content commissioned by the advertiser, and obtain user preference characteristics and advertising characteristics based on the analysis results.

[0010] According to the above technical solution, the advertising delivery module also includes a delivery analysis module, which is used to analyze the browsing data and browsing time of users at various web page levels on the web page side, analyze the advertiser's budget, allocate the budget, formulate an advertising delivery plan based on the allocation result, analyze the historical benefits of advertising, train the prediction model, analyze the matching degree between users and advertisements, market changes and strategy changes, and revise the advertising benefit prediction results based on the analysis results.

[0011] According to the above technical solution, the advertisement design module includes a communication module and an effect display module. The communication module is used to communicate with advertisers to adjust the advertisement design solution, and the effect display module is used to render the advertisement and display it to the advertiser.

[0012] According to the above technical solution, the operation method of the advertising design platform mainly includes the following steps:

[0013] Step S1: Collect users' comprehensive browsing data, users' feedback data and purchase records through the data collection module, and enter historical advertising comprehensive benefit data into the system through the historical data entry module;

[0014] Step S2: After the data is entered into the system, the system starts the bias analysis module, starts analyzing the user's comprehensive browsing data and the advertising content, obtains the user's preferences and advertising features based on the analysis results, and further matches the user with the advertisement;

[0015] Step S3: During the advertisement delivery process, the system starts the delivery analysis module, starts to analyze the user's consumption willingness, consumption level, consumption time, browsing data and browsing time at each web page level in the web page, analyzes the advertiser's budget, allocates the budget to formulate an advertisement delivery plan, analyzes the effectiveness of the advertisement plan, and adjusts the advertisement delivery plan according to the analysis results;

[0016] Step S4: When designing an advertisement, real-time communication is performed with the advertiser through the universal module, the advertisement is rendered, and the rendering result is displayed on the advertiser's terminal.

[0017] According to the above technical solution, step S2 further includes the following steps:

[0018] Step S21: Obtain comprehensive browsing data of the user, identify keywords of the user browsing data and keywords of the search data, compare the keywords of the user browsing data and the keywords of the search data, if the similarity is greater than a maximum threshold, merge the keywords and mark them as the first user, if the similarity is less than the maximum threshold and greater than the minimum threshold, weightedly merge the keywords and mark them as the second user, otherwise mark them as the third user, and retrieve the corresponding user preferences in the database according to the keywords;

[0019] Step S22: Acquire the advertising commission content of the advertiser, identify the advertising-oriented keywords in the advertising commission content, and retrieve the corresponding advertising features in the database according to the keywords;

[0020] Step S23: Obtain advertising features and user preferences, retrieve the advertising features and user preference matching knowledge graph in the database, match the advertising features with the user preferences according to the knowledge graph, and if the matching degree is greater than the first threshold, mark it as the target user, otherwise the system continues to detect.

[0021] According to the above technical solution, step S3 further includes the following steps:

[0022] Step S31: Obtain the user's purchase record on the web page, identify the characteristics of the goods purchased by the user and the corresponding quantity, and calculate the proportion of the goods corresponding to the characteristics purchased by the user through a formula , where Indicates the proportion of users who purchase products corresponding to the product features. Indicates the number of products that users purchase corresponding to the product feature. Indicates the total number of products purchased by the user. If the proportion of products corresponding to the product feature purchased by the user is greater than the threshold, the user is marked in the database as willing to purchase the product corresponding to the product feature. Otherwise, the system continues to detect.

[0023] Step S32: Identify the total amount of money spent by the user on the goods corresponding to the goods feature during the period, and calculate the proportion of money spent by the user on the goods corresponding to the goods feature during the period through a formula. If the proportion of money spent on the goods corresponding to the goods feature is greater than a threshold, mark the user's consumption level as meeting the requirement in the database, otherwise the system continues to detect;

[0024] Step S33: Identify the timestamps corresponding to the user's purchase records within the period, calculate the time interval between two adjacent timestamps, and if the time interval is less than the threshold, mark the timestamps at both ends of the time interval; otherwise, the system continues to detect and identify the marks in the timestamps. If there are marks in adjacent timestamps, connect the timestamps; otherwise, continue to detect, identify the position of the connected time period within the period, and mark the time period.

[0025] According to the above technical solution, step S33 further includes the following steps:

[0026] Step S331: Acquire the target user, identify the consumption intention mark in the target user, and when the target user has consumption intention, retrieve the commodity features corresponding to the consumption intention in the database. If the matching degree between the commodity features and the target user preference is greater than the threshold, the system continues to detect, otherwise the target user is deleted, and identify the consumption level mark in the target user. If the consumption level of the target user is less than the threshold, the target user is deleted, otherwise the system continues to detect;

[0027] Step S332: obtaining the URL of the web page level browsed by the target user on the web page, retrieving the corresponding browsing time in the database according to the URL of the web page level, sorting the browsing time in descending order, and selecting the first-ranked web page level URL for marking;

[0028] Step S333: Obtain the daily advertising budget entrusted by the advertiser and the number of target users, retrieve the corresponding rated daily budget in the database according to the number of target users, retrieve the corresponding first delivery plan in the database according to the daily budget, calculate the difference between the daily advertising budget entrusted by the advertiser and the rated daily budget in the database, if the difference is less than the threshold, identify the matching degree between the user preferences and the advertising features of the target users, sort the users in descending order according to the matching degree, delete the users with a ranking less than the threshold, and deliver the advertisement according to the first delivery plan, otherwise obtain the remaining target users in the database, identify the web page level URL mark in the target users, anchor the web page level URL mark position in the target users, deliver the advertisement to the URL, identify the consumption time mark of the target users, when the current date overlaps with the user consumption time mark, obtain all the web page URLs browsed by the target users, deliver the advertisement to all the web page URLs browsed by the target users, and mark the plan as the second plan.

[0029] According to the above technical solution, step S333 further includes the following steps:

[0030] Step A1: Obtain historical advertising comprehensive benefit data, pre-process the historical advertising comprehensive benefit data, identify the advertising benefit characteristics of the processed data, use the processed data and advertising benefit characteristics as sample data, use the sample data to train the advertising benefit prediction model, and predict the current advertising benefit based on the advertising benefit model ;

[0031] Step A2: Obtain advertisement features, retrieve corresponding market data from the database based on the advertisement features, identify the supply and demand relationship in the market, and retrieve the corresponding coefficient of influence on advertisement benefits from the database based on the supply and demand relationship in the market , identify the matching degree between user preferences and advertising features, and retrieve the corresponding advertising benefit impact coefficient in the database according to the matching degree , identify the tags in the target user, and retrieve the influence weight of the corresponding coefficient of influence on the advertising benefit according to the first, second and third tags ;

[0032] Step A3: Calculate the corrected advertising benefits through the formula , calculate the difference between the corrected advertising benefit and the predicted advertising benefit. If the difference is greater than the threshold, the plan is marked as the target plan. Otherwise, the advertising plan is deleted and the system delivers advertising according to the target plan.

[0033] Compared with the prior art, the beneficial effects achieved by the present invention are as follows: the present invention can accurately analyze the user's purchasing tendency by analyzing the user's browsing data, search data and the advertising content commissioned by the advertiser, thereby more accurately matching advertisements, greatly improving the accuracy of the system; by analyzing the user's consumption level, that is, purchasing power, it can avoid the user's insufficient purchasing power, resulting in low advertising revenue, thereby greatly improving the advertising revenue of the system; by analyzing the user's spending at various time points within a cycle, it can accurately deliver advertisements in the interval where the user has the purchasing power, further improving the accuracy of the system; by analyzing the web page levels that the target user frequently browses, the target user's browsing preference is obtained, and then the advertising delivery plan is adjusted according to the user's browsing preference and the user's consumption time, which can ensure that the user can see the advertisement when there is purchasing power, promote the target user to place an order, thereby greatly improving the advertising revenue; by analyzing the market situation and the impact of the matching degree of user preferences and advertising characteristics on advertising effectiveness, it can more accurately predict the benefits of the advertising delivery plan, thereby avoiding the low benefits of the advertising delivery plan, further improving the accuracy of the system and the benefits of advertising. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0035] Figure 1 It is a schematic diagram of the system module composition of the present invention. DETAILED DESCRIPTION

[0036] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0037] See also Figure 1The present invention provides a technical solution: a multimedia advertisement design platform based on the Internet of Things, including a data collection module, an advertisement delivery module and an advertisement design module. The data collection module is used to collect comprehensive browsing data of users, feedback data of users and purchase records, and input historical advertisement comprehensive benefit data into the system. The advertisement delivery module is used to analyze the user's preference for browsing information on the web page, analyze the user's consumption willingness and consumption level, analyze the characteristics of the advertisement content entrusted by the advertiser, match the user with the advertisement according to the analysis result, analyze the browsing level preferred by the user, formulate an advertisement delivery plan according to the analysis result and the matching result, further analyze the revenue effect of the delivery plan, and determine the advertisement delivery plan according to the revenue effect. The advertisement design module is used to render the advertisement effect and display it to the advertiser, communicate with the advertiser through the communication module to modify the advertisement design, and the data collection module, the advertisement delivery module and the advertisement design module are connected to each other in communication;

[0038] The bias analysis module includes a user preference analysis submodule and an advertisement feature analysis submodule. The user preference analysis submodule is used to analyze the browsing data, consumption willingness and consumption level of users on the web page. The advertisement feature analysis submodule is used to analyze the advertisement content commissioned by the advertiser and obtain the user preference features and advertisement features according to the analysis results.

[0039] The delivery analysis module includes a delivery level analysis submodule, a delivery plan formulation submodule and a benefit analysis submodule. The delivery level analysis submodule is used to analyze the user's browsing data and corresponding browsing time at each web page level, and obtain the user's browsing characteristics based on the analysis results. The delivery plan formulation submodule is used to analyze the advertiser's advertising budget, allocate the budget and formulate an advertising delivery plan. The benefit analysis submodule is used to analyze historical advertising benefits, train prediction models, analyze the matching degree between users and advertisements, market changes and strategy changes, and revise the advertising benefit prediction results based on the analysis results.

[0040] The data collection module includes a data acquisition module and a historical data entry module. The data acquisition module is used to collect users' comprehensive browsing data, users' feedback data and purchase records, and the historical data entry module is used to enter historical advertising comprehensive benefit data into the system.

[0041] The advertisement delivery module includes a preference analysis module, which is used to analyze the user's browsing data, search data, consumption intention and consumption level on the web page, analyze the advertisement content commissioned by the advertiser, and obtain user preference characteristics and advertisement characteristics based on the analysis results.

[0042] The advertising delivery module also includes a delivery analysis module, which is used to analyze the browsing data and browsing time of users at various web page levels on the web page, analyze the advertiser's budget, allocate the budget, formulate an advertising delivery plan based on the allocation results, analyze the historical effectiveness of advertising, train prediction models, analyze the matching degree between users and advertisements, market changes and strategy changes, and revise the advertising effectiveness prediction results based on the analysis results.

[0043] The advertisement design module includes a communication module and an effect display module. The communication module is used to communicate with advertisers to adjust the advertisement design plan, and the effect display module is used to render the advertisement and display it to the advertiser.

[0044] The operation method of the advertising design platform mainly includes the following steps:

[0045] Step S1: Collect users' comprehensive browsing data, users' feedback data and purchase records through the data collection module, and enter historical advertising comprehensive benefit data into the system through the historical data entry module;

[0046] Step S2: After the data is entered into the system, the system starts the bias analysis module, starts analyzing the user's comprehensive browsing data and the advertising content, obtains the user's preferences and advertising features based on the analysis results, and further matches the user with the advertisement;

[0047] Step S3: During the advertisement delivery process, the system starts the delivery analysis module, starts to analyze the user's consumption willingness, consumption level, consumption time, browsing data and browsing time at each web page level in the web page, analyzes the advertiser's budget, allocates the budget to formulate an advertisement delivery plan, analyzes the effectiveness of the advertisement plan, and adjusts the advertisement delivery plan according to the analysis results;

[0048] Step S4: When designing an advertisement, real-time communication is performed with the advertiser through the universal module, the advertisement is rendered, and the rendering result is displayed on the advertiser's terminal.

[0049] Step S2 further comprises the following steps:

[0050] Step S21: Obtain comprehensive browsing data of the user, identify keywords of the user browsing data and keywords of the search data, compare the keywords of the user browsing data and the keywords of the search data, if the similarity is greater than a maximum threshold, merge the keywords and mark them as the first user, if the similarity is less than the maximum threshold and greater than the minimum threshold, weightedly merge the keywords and mark them as the second user, otherwise mark them as the third user, and retrieve the corresponding user preferences in the database according to the keywords;

[0051] Step S22: Acquire the advertising commission content of the advertiser, identify the advertising-oriented keywords in the advertising commission content, and retrieve the corresponding advertising features in the database according to the keywords;

[0052] Step S23: Obtain advertising features and user preferences, retrieve the advertising features and user preference matching knowledge graph in the database, match the advertising features with the user preferences according to the knowledge graph, and if the matching degree is greater than the first threshold, mark it as the target user, otherwise the system continues to detect. By analyzing the user's browsing data, retrieval data and the advertising content commissioned by the advertiser, it can accurately analyze the user's purchasing tendencies, and then more accurately match advertisements, greatly improving the accuracy of the system.

[0053] Step S3 further comprises the following steps:

[0054] Step S31: Obtain the user's purchase record on the web page, identify the characteristics of the goods purchased by the user and the corresponding quantity, and calculate the proportion of the goods corresponding to the characteristics purchased by the user through a formula , where Indicates the proportion of users who purchase products corresponding to the product features. Indicates the number of products that users purchase corresponding to the product feature. Indicates the total number of products purchased by the user. If the proportion of products corresponding to the product feature purchased by the user is greater than the threshold, the user is marked in the database as willing to purchase the product corresponding to the product feature. Otherwise, the system continues to detect.

[0055] Step S32: Identify the total expenditure of the user on the goods corresponding to the commodity feature during the period, and calculate the proportion of the expenditure of the user on the goods corresponding to the commodity feature during the period through a formula. If the proportion of the expenditure on the goods corresponding to the commodity feature is greater than the threshold, the user's consumption level is marked in the database as meeting the requirement. Otherwise, the system continues to detect. By analyzing the user's consumption level, that is, purchasing power, it is possible to avoid insufficient purchasing power of the user, resulting in low advertising revenue, thereby greatly improving the advertising revenue of the system;

[0056] Step S33: Identify the timestamps corresponding to the user's purchase records within the cycle, calculate the time interval between two adjacent timestamps, and if the time interval is less than the threshold, mark the timestamps at both ends of the time interval; otherwise, the system continues to detect and identify the marks in the timestamps. If there are marks in adjacent timestamps, connect the timestamps; otherwise, continue to detect and identify the position of the connected time period within the cycle, and mark the time period. By analyzing the user's spending at each time point within the cycle, accurate advertising can be delivered in the intervals where the user has the purchasing power, further improving the accuracy of the system.

[0057] Step S33 further includes the following steps:

[0058] Step S331: Acquire the target user, identify the consumption intention mark in the target user, and when the target user has consumption intention, retrieve the commodity features corresponding to the consumption intention in the database. If the matching degree between the commodity features and the target user preference is greater than the threshold, the system continues to detect, otherwise the target user is deleted, and identify the consumption level mark in the target user. If the consumption level of the target user is less than the threshold, the target user is deleted, otherwise the system continues to detect;

[0059] Step S332: obtaining the URL of the web page level browsed by the target user on the web page, retrieving the corresponding browsing time in the database according to the URL of the web page level, sorting the browsing time in descending order, and selecting the first-ranked web page level URL for marking;

[0060] Step S333: Obtain the advertiser's commissioned advertising daily budget and the number of target users. According to the number of target users, retrieve the corresponding rated daily budget in the database. According to the daily budget, retrieve the corresponding first delivery plan in the database. Calculate the difference between the advertiser's commissioned advertising daily budget and the rated daily budget in the database. If the difference is less than the threshold, identify the matching degree between the user preferences and the advertising features of the target users. According to the matching degree, sort the users in descending order, delete the users whose ranking is less than the threshold, and deliver the advertisements according to the first delivery plan. Otherwise, obtain the remaining target users in the database, identify the web page-level URL tags in the target users, and anchor the target users. The web page level URL mark position in the user's browser is used to place the advertisement on the URL, and the consumption time mark of the target user is identified. When the current date overlaps with the user's consumption time mark, all the web page URLs browsed by the target user are obtained, and the advertisement is placed on all the web page URLs browsed by the target user. This scheme is marked as the second scheme. By analyzing the web page levels frequently browsed by the target user, the browsing preference of the target user is obtained, and then the advertisement delivery scheme is adjusted according to the user's browsing preference and the user's consumption time. This can ensure that the user can see the advertisement when he has the purchasing power, promote the target user to place an order, and thus greatly increase the advertising revenue.

[0061] Step S333 further includes the following steps:

[0062] Step A1: Obtain historical advertising comprehensive benefit data, pre-process the historical advertising comprehensive benefit data, identify the advertising benefit characteristics of the processed data, use the processed data and advertising benefit characteristics as sample data, use the sample data to train the advertising benefit prediction model, and predict the current advertising benefit based on the advertising benefit model ;

[0063] Step A2: Obtain advertisement features, retrieve corresponding market data from the database based on the advertisement features, identify the supply and demand relationship in the market, and retrieve the corresponding coefficient of influence on advertisement benefits from the database based on the supply and demand relationship in the market , identify the matching degree between user preferences and advertising features, and retrieve the corresponding advertising benefit impact coefficient in the database according to the matching degree , identify the tags in the target user, and retrieve the influence weight of the corresponding coefficient of influence on the advertising benefit according to the first, second and third tags ;

[0064] Step A3: Calculate the corrected advertising benefits through the formula , calculate the difference between the corrected advertising benefit and the predicted advertising benefit. If the difference is greater than the threshold, the plan is marked as the target plan, otherwise the advertising plan is deleted. The system delivers advertisements according to the target plan. By analyzing the impact of market conditions and the matching degree of user preferences and advertising features on advertising benefits, it can more accurately predict the benefits of advertising plans, thereby avoiding low benefits of advertising plans, and further improving the accuracy of the system and the benefits of advertising.

[0065] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0066] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments or replace some of the technical features therein by equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A multimedia advertising design platform based on the Internet of Things, including a data collection module, an advertising delivery module and an advertising design module, characterized in that: The data collection module is used to collect users' comprehensive browsing data, users' feedback data and purchase records, and enter historical advertising comprehensive benefit data into the system. The advertising delivery module is used to analyze users' preferences for browsing information on web pages, analyze users' consumption willingness and consumption level, analyze the content characteristics of advertisements commissioned by advertisers, match users with advertisements according to the analysis results, analyze users' preferred browsing levels, formulate advertising delivery plans according to the analysis results and matching results, further analyze the revenue effect of the delivery plans, and determine the advertising delivery plans according to the revenue effect. The advertising design module is used to render advertising effects and display them to advertisers, communicate with advertisers through the communication module to modify advertising designs, and the data collection module, advertising delivery module and advertising design module are connected to each other for communication; The advertisement delivery module includes a bias analysis module, which is used to analyze the browsing data, search data, consumption intention and consumption level of users on the web page, analyze the advertisement content entrusted by the advertiser, and obtain user preference characteristics and advertisement characteristics according to the analysis results; The advertising delivery module also includes a delivery analysis module, which is used to analyze the browsing data and browsing time of users at various web page levels on the web page end, analyze the advertiser's budget, allocate the budget, formulate an advertising delivery plan based on the allocation result, analyze the historical advertising benefits, train the prediction model, analyze the matching degree between users and advertisements, market changes and strategy changes, and revise the advertising benefit prediction results based on the analysis results; The bias analysis module includes a user preference analysis submodule and an advertisement feature analysis submodule. The user preference analysis submodule is used to analyze the browsing data, consumption willingness and consumption level of users on the web page. The advertisement feature analysis submodule is used to analyze the advertisement content commissioned by the advertiser, and obtain the user preference features and advertisement features according to the analysis results. The delivery analysis module includes a delivery level analysis submodule, a delivery plan formulation submodule and a benefit analysis submodule. The delivery level analysis submodule is used to analyze the browsing data and corresponding browsing time of users at each web page level, and obtain the browsing characteristics of users according to the analysis results. The delivery plan formulation submodule is used to analyze the budget of advertisers for advertising, allocate the budget and formulate an advertising delivery plan. The benefit analysis submodule is used to analyze historical advertising benefits, train prediction models, analyze the matching degree between users and advertisements, market changes and strategy changes, and revise the advertising benefit prediction results according to the analysis results. The data collection module includes a data collection module and a historical data entry module. The data collection module is used to collect users' comprehensive browsing data, users' feedback data and purchase records. The historical data entry module is used to enter historical advertising comprehensive benefit data into the system. The advertisement design module includes a communication module and an effect display module, wherein the communication module is used to communicate with the advertiser to adjust the advertisement design plan, and the effect display module is used to render the advertisement and display it to the advertiser; The operation method of the advertising design platform mainly includes the following steps: Step S1: Collect users' comprehensive browsing data, users' feedback data and purchase records through the data collection module, and enter historical advertising comprehensive benefit data into the system through the historical data entry module; Step S2: After the data is entered into the system, the system starts the bias analysis module, starts analyzing the user's comprehensive browsing data and the advertising content, obtains the user's preferences and advertising features based on the analysis results, and further matches the user with the advertisement; Step S3: During the advertisement delivery process, the system starts the delivery analysis module, starts to analyze the user's consumption willingness, consumption level, consumption time, browsing data and browsing time at each web page level in the web page, analyzes the advertiser's budget, allocates the budget to formulate an advertisement delivery plan, analyzes the effectiveness of the advertisement plan, and adjusts the advertisement delivery plan according to the analysis results; Step S4: When designing an advertisement, communicate with the advertiser in real time through the universal module, render the advertisement, and display the rendering result on the advertiser's terminal; The step S2 further comprises the following steps: Step S21: Obtain comprehensive browsing data of the user, identify keywords of the user browsing data and keywords of the search data, compare the keywords of the user browsing data and the keywords of the search data, if the similarity is greater than a maximum threshold, merge the keywords and mark them as the first user, if the similarity is less than the maximum threshold and greater than the minimum threshold, weightedly merge the keywords and mark them as the second user, otherwise mark them as the third user, and retrieve the corresponding user preferences in the database according to the keywords; Step S22: Acquire the advertising commission content of the advertiser, identify the advertising-oriented keywords in the advertising commission content, and retrieve the corresponding advertising features in the database according to the keywords; Step S23: Obtain advertisement features and user preferences, retrieve advertisement features and user preference matching knowledge graphs in the database, match advertisement features with user preferences according to the knowledge graphs, and if the matching degree is greater than a first threshold, mark the user as a target user, otherwise the system continues to detect; The step S3 further comprises the following steps: Step S31: Obtain the user's purchase record on the web page, identify the characteristics of the goods purchased by the user and the corresponding quantity, and calculate the proportion of the goods corresponding to the characteristics purchased by the user through a formula , where Indicates the proportion of users who purchase products corresponding to the product features. Indicates the number of products that users purchase corresponding to the product feature. Indicates the total number of products purchased by the user. If the proportion of products corresponding to the product feature purchased by the user is greater than the threshold, the user is marked in the database as willing to purchase the product corresponding to the product feature. Otherwise, the system continues to detect. Step S32: Identify the total amount of money spent by the user on the goods corresponding to the goods feature during the period, and calculate the proportion of money spent by the user on the goods corresponding to the goods feature during the period through a formula. If the proportion of money spent on the goods corresponding to the goods feature is greater than a threshold, mark the user's consumption level as meeting the requirement in the database, otherwise the system continues to detect; Step S33: Identify the timestamps corresponding to the user's purchase records within the period, calculate the time interval between two adjacent timestamps, and if the time interval is less than the threshold, mark the timestamps at both ends of the time interval; otherwise, the system continues to detect and identify the marks in the timestamps. If there are marks in the adjacent timestamps, connect the timestamps; otherwise, continue to detect, identify the position of the connected time period within the period, and mark the time period; The step S33 further comprises the following steps: Step S331: Acquire the target user, identify the consumption intention mark in the target user, and when the target user has consumption intention, retrieve the commodity features corresponding to the consumption intention in the database. If the matching degree between the commodity features and the target user preference is greater than the threshold, the system continues to detect, otherwise the target user is deleted, and identify the consumption level mark in the target user. If the consumption level of the target user is less than the threshold, the target user is deleted, otherwise the system continues to detect; Step S332: obtaining the URL of the web page level browsed by the target user on the web page, retrieving the corresponding browsing time in the database according to the URL of the web page level, sorting the browsing time in descending order, and selecting the first-ranked web page level URL for marking; Step S333: Obtain the daily budget of the advertisement entrusted by the advertiser and the number of target users, retrieve the corresponding rated daily budget in the database according to the number of target users, retrieve the corresponding first delivery plan in the database according to the daily budget, calculate the difference between the daily budget of the advertisement entrusted by the advertiser and the rated daily budget in the database, if the difference is less than the threshold, identify the matching degree between the user preference and the advertisement feature of the target users, sort the users in descending order according to the matching degree, delete the users whose ranking is less than the threshold, and deliver the advertisement according to the first delivery plan, otherwise obtain the remaining target users in the database, identify the web page level URL mark in the target users, anchor the web page level URL mark position in the target users, deliver the advertisement to the URL, identify the consumption time mark of the target users, when the current date overlaps with the user consumption time mark, obtain all the web page URLs browsed by the target users, deliver the advertisement to all the web page URLs browsed by the target users, and mark the plan as the second plan; The step S333 further includes the following steps: Step A1: Obtain historical advertising comprehensive benefit data, pre-process the historical advertising comprehensive benefit data, identify the advertising benefit characteristics of the processed data, use the processed data and advertising benefit characteristics as sample data, use the sample data to train the advertising benefit prediction model, and predict the current advertising benefit based on the advertising benefit model ; Step A2: Obtain advertisement features, retrieve corresponding market data from the database based on the advertisement features, identify the supply and demand relationship in the market, and retrieve the corresponding coefficient of influence on advertisement benefits from the database based on the supply and demand relationship in the market , identify the matching degree between user preferences and advertising features, and retrieve the corresponding advertising benefit impact coefficient in the database according to the matching degree , identify the tags in the target user, and retrieve the influence weight of the corresponding coefficient of influence on the advertising benefit according to the first, second and third tags ; Step A3: Calculate the corrected advertising benefits through the formula , calculate the difference between the corrected advertising benefit and the predicted advertising benefit. If the difference is greater than the threshold, the plan is marked as the target plan. Otherwise, the advertising plan is deleted and the system delivers advertising according to the target plan.

Images