Personalized advertisement content accurate delivery method and system based on user portraits

Abstract

The invention discloses a personalized advertisement content accurate delivery method and system based on user portraits, and belongs to the technical field of intelligent popularization. According to the method, user portraits such as gender / age and elevator scene real-time data are combined, an advertisement playing benefit score R is dynamically calculated, accurate matching of advertisement content and current audiences is achieved, and the user portraits are accurately delivered. And the conversion efficiency of the single advertisement is improved. In addition, a double evaluation mechanism of benefit score sorting and advertisement importance is adopted, and an optimal playing scheme is dynamically generated through an H1 value minimization algorithm, so that the exposure rights and interests of high-value advertisements are guaranteed, and a playing strategy can be optimized in real time according to historical effects. According to the invention, the progress difference of each advertisement is balanced by introducing and completing proportion standard deviation monitoring and dynamically adjusting the playing sequence. And long-term overstock of part of advertisements due to low benefit score is effectively avoided.

Description

Technical Field

[0001] The present invention belongs to the field of intelligent promotion technology, and specifically relates to a method and system for accurately delivering personalized advertising content based on user portraits. Background Art

[0002] Since its inception in the 1990s, elevator advertising has gone through three major stages of technological development: Static print advertising: Mainly paper or framed posters, it relies on limited labels such as building type and traffic flow time for delivery, and has problems such as extensive delivery, difficult to quantify results, and content homogeneity. LCD advertising outside elevators: Companies such as Focus Media have introduced dynamic video ads to enhance visual impact, but are still limited by the short waiting time outside elevators (usually <10 seconds) and cannot accurately match user attributes. Smart screen advertising in elevators: Companies such as Xinchao Media use digital screens that support dynamic content updates, AI visual recognition, time-sharing delivery, and interactive functions, gradually replacing traditional frame ads and outside elevators.

[0003] Although smart screen technology has significantly improved the efficiency of advertising delivery, the following technical bottlenecks still exist: the lack of dynamic analysis of the real-time behavior of elevator passengers leads to a low accuracy rate in interest prediction. The delivery model still wastes resources: the traditional CPD (pay-per-day) model easily leads to "empty elevator broadcasts", and the content generation and scene adaptability are poor: the advertising content is mostly fixed templates and cannot be adjusted in real time according to the characteristics of the elevator passengers (such as age and gender). For example, when maternal and child advertisements are delivered in elderly communities, the conversion rate drops by more than 40%. In order to solve the above problems, the present invention provides the following technical solutions. Summary of the Invention

[0004] The purpose of the present invention is to provide a method and system for precise delivery of personalized advertising content based on user portraits, so as to solve the problems of poor flexibility and low conversion rate in the existing technology of advertising delivery.

[0005] The purpose of the present invention can be achieved through the following technical solutions:

[0006] The method for accurately delivering personalized advertising content based on user portraits includes the following steps:

[0007] The first step is to confirm the target audience of each advertisement and the corresponding profile data;

[0008] Get the proportion of each audience type corresponding to each advertisement;

[0009] The value of i ranges from 1 to n, where n represents the number of audience types;

[0010] The second step is to collect image information inside the elevator at time t1 before the next advertisement is played, using the image acquisition unit inside the elevator; where t1 is a preset value;

[0011] Analyze and obtain the gender and age of each person in the elevator;

[0012] Divide the people in the elevator into several character groups based on their gender and age, and each character group corresponds to a target audience type;

[0013] Obtain each advertisement that can be played by the display unit and record it as an advertisement to be played;

[0014] Obtaining the advertisement playing benefit score R corresponding to the composition of people in the current elevator when assuming that each advertisement to be played is played;

[0015] In the third step, the advertisement to be played with the largest corresponding advertisement playing benefit score R is selected as the advertisement to be played next.

[0016] As a further solution of the present invention, when determining the number of audience groups, the proportion of each audience group is sorted in descending order, and then added in descending order until the comprehensive proportion exceeds the preset proportional coefficient. The audience groups involved in the summation calculation are used as the final confirmed audience group types for the advertisement, and the proportions are normalized.

[0017] As a further solution of the present invention, the calculation method of the advertisement broadcast benefit score R is as follows:

[0018] Among them, ai represents the corresponding proportion of each character group, and bi represents the number of people in each character group in the elevator.

[0019] As a further solution of the present invention, an advertisement playing scheme optimization strategy is also included, including the following steps:

[0020] Obtaining the advertising play benefit score R of each advertisement played by each display unit within a preset time range in the past;

[0021] For a display unit, sort the advertisements played by it in descending order according to the sum of the corresponding advertisement play benefit scores R, which is expressed as ek;

[0022] Where k represents the order of each advertisement, and ek represents the sum of the specific advertisement broadcast benefit scores R;

[0023] Sort the ads by importance, and then determine the display units for each ad in that order to form several sets of plans.

[0024] For one of the solutions, after all advertisements have completed the determination of display units, the order of each advertisement in the corresponding display units is calculated, and then the sum H of the orders corresponding to each advertisement is calculated;

[0025] Calculate the sum H1 of the sum H of the corresponding orders of all advertisements;

[0026] The H1 value corresponding to each plan is calculated in turn, and the plan with the smallest H1 is determined as the final advertising broadcast plan, and the advertisement is broadcast according to the advertising broadcast plan in the next cycle.

[0027] As a further solution of the present invention, when the number of advertisements to be played changes:

[0028] If the number of advertisements is reduced on the existing basis, delete the corresponding advertisements in the advertisement playing plan;

[0029] When the number increases, the added advertisements are trial-played in each display unit and then the advertisement play plan optimization strategy is executed based on the play results.

[0030] As a further solution of the present invention, when the number of advertisements to be played and the required playback volume of each advertisement are determined, the playback order of each advertisement is further allocated. The specific steps are as follows:

[0031] Calculate the advertising benefit score R of each advertisement to be played in sequence;

[0032] Get the completion ratio cj of each advertisement to be played in the current cycle; j ranges from 1 to m, where m represents the number of advertisements to be played;

[0033] The completion ratio refers to the sum of the advertising play benefit scores R corresponding to each play of the advertisement to be played in the current cycle / the sum of the advertising play benefit scores R required to be completed in the current cycle;

[0034] Calculate the completion ratio cj of each pending ad before a new pending ad is played, and then calculate the standard deviation S1 of these m cj values;

[0035] Calculate the standard deviation S2 of the advertisements to be played after the completion ratio cj is updated after playing each advertisement to be played;

[0036] Calculate m corresponding |S1-S2| values, select the advertisement to be played corresponding to the S2 value with the smallest |S1-S2| value, and use the advertisement to be played as the next advertisement to be played by the display unit.

[0037] As a further solution of the present invention, a method for adjusting the number of advertisements to be played on each display unit is also included, which is specifically as follows:

[0038] When a display unit completes playing of advertisements to be played within a cycle;

[0039] First, check whether there is time left in the cycle. If so, add the number of ads to be played in the display unit.

[0040] If not:

[0041] For an advertisement, obtain the order g of its corresponding advertisement playing benefits R when it is sorted by size each time it is used as an advertisement to be played;

[0042] Get the average value G of the order g corresponding to each playback of the advertisement in the entire cycle;

[0043] Obtain an average value Gp of G values ​​of all advertisements to be played after a display unit completes playing of advertisements to be played within a cycle;

[0044] When the average value Gp is ​​greater than the preset value Gy, the number of advertisements to be played is increased in the display unit.

[0045] The present invention also discloses a personalized advertising content precision delivery system based on user portraits, which is used to execute the above delivery method, and includes:

[0046] An image acquisition unit is used to acquire scene images in the elevator and transmit the acquired scene image information to the image analysis unit;

[0047] An image analysis unit, configured to analyze scene images and identify the number, gender, and age of people therein;

[0048] The main control unit is used to determine the next advertisement to be played based on the portrait data of the people in the elevator analyzed by the image analysis unit;

[0049] Display unit, used for playing advertisements.

[0050] Beneficial effects of the present invention:

[0051] The present invention combines user portrait features such as gender / age with real-time data of elevator scenes, and dynamically calculates the advertising playback benefit score R to achieve accurate matching of advertising content with the current audience, thereby improving the conversion efficiency of a single advertisement.

[0052] The present invention adopts a dual evaluation mechanism of benefit score ranking and advertising importance, and dynamically generates the optimal playback plan through the H1 value minimization algorithm, which not only guarantees the exposure rights of high-value advertisements, but also optimizes the playback strategy in real time according to historical effects.

[0053] This method uses the standard deviation monitoring of completion rates (minimizing the |S1-S2| principle) to dynamically adjust the playback order to balance progress differences among ads. Experimental data shows that this method can reduce ad progress deviation by approximately 37% (tested on a weekly basis), effectively preventing some ads from being held in a backlog due to low effectiveness scores.

[0054] This invention intelligently expands the advertising capacity of inefficient display units by comparing the average advertising benefit order Gp with the threshold Gy. Practical applications have shown that this strategy can increase advertising revenue on a single elevator screen by 15-22%.

[0055] Through intelligent decision-making in the dual dimensions of time and space, the present invention achieves the optimal balance between precise delivery, dynamic optimization and resource utilization in elevator advertising scenarios, and is particularly suitable for mining advertising value in highly mobile enclosed spaces. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] The present invention will be further described below with reference to the accompanying drawings.

[0057] Figure 1 It is a flow chart of a method for precise delivery of personalized advertising content based on user portraits according to an embodiment of the present invention. DETAILED DESCRIPTION

[0058] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. 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 making any creative efforts shall fall within the scope of protection of the present invention.

[0059] Example 1

[0060] A personalized advertising content precision delivery method based on user portraits, such as Figure 1 As shown, the following steps are included:

[0061] The first step is to identify the target audience for each ad, as well as the profile data for each type of audience.

[0062] Get the proportion of each audience type corresponding to each advertisement;

[0063] The value of i ranges from 1 to n, where n represents the number of audience types;

[0064] Since the target audiences of many products are relatively complex, in addition to the main target audiences, there are also many other smaller target audiences. Therefore, in order to simplify the calculation, the proportions of each target audience can be sorted from largest to smallest, and then added up in descending order until the combined proportion exceeds the preset ratio coefficient. The target audiences included in the summation calculation are used as the final confirmed target audience types for the advertisement, and the proportions of each target audience type for the final confirmed advertisement are proportionally expanded so that the sum of the proportions of each target audience type for the final confirmed advertisement is 1.

[0065] The audience profile data here includes gender and age;

[0066] Specifically, the audience profile of the product corresponding to the advertisement can be obtained through purchase records, questionnaire surveys, etc. This method of obtaining audience profiles is a commonly used well-known technology in research, so it will not be specifically explained or limited here;

[0067] The second step is to collect image information inside the elevator at time t1 before the next advertisement is played, using the image acquisition unit inside the elevator; where t1 is a preset value;

[0068] The image analysis unit analyzes the image information collected by the image acquisition unit to obtain the gender and age of each person in the elevator;

[0069] Classify the people in the elevator according to their gender and age, and divide them into several character groups. Each character group corresponds to a target audience type.

[0070] Obtain each advertisement that can be played by the display unit, and mark these advertisements as pending advertisements for clarity of subsequent descriptions;

[0071] Obtaining the advertisement playing benefit score R corresponding to the composition of people in the current elevator when assuming that each advertisement to be played is played;

[0072] The calculation method of advertising play benefit score R is:

[0073] Where ai represents the proportion of each character group, and bi represents the number of people in each character group in the elevator;

[0074] After calculation according to the above method, the larger the advertising playback benefit score R of an advertisement to be played, the higher the playback effect and revenue.

[0075] In the third step, the advertisement with the largest corresponding advertisement playing benefit score R is selected as the next advertisement to be played. This method can make full use of traffic resources and maximize the benefit of advertisement playing.

[0076] Step 4: Obtain the advertising play benefit score R of each advertisement played by each display unit within the past preset time range;

[0077] For a display unit, sort the advertisements played by it in descending order according to the sum of the corresponding advertisement play benefit scores R, which is expressed as ek;

[0078] Where k represents the order of each advertisement, and ek represents the sum of the specific advertisement broadcast benefit scores R;

[0079] The fifth step is to sort the ads in descending order based on their importance scores, and then determine the display unit for each ad in that order.

[0080] Depending on the choices made, multiple plans will be formed;

[0081] For one of the solutions, after all advertisements have completed the determination of display units, the order of each advertisement in the corresponding display units is calculated, and then the sum H of the orders corresponding to each advertisement is calculated;

[0082] Calculate the sum H1 of the sum H of the corresponding orders of all advertisements;

[0083] The H1 value corresponding to each plan is calculated in turn, and the plan with the smallest H1 is determined as the final advertising broadcast plan, and the advertisement is broadcast according to the advertising broadcast plan in the next cycle.

[0084] The importance score of the advertisement may be determined based on the revenue of the advertisement, and the higher the revenue, the higher the corresponding score.

[0085] When the number of ads to be played changes:

[0086] If the number of advertisements is reduced on the existing basis, delete the corresponding advertisements in the advertisement playing plan;

[0087] When the number increases, the added advertisements are trial-played in each display unit (ie, they are added to the advertisements to be played in each display unit) and then the operations in the fourth and fifth steps are performed according to the playback results.

[0088] Example 2

[0089] Since, in actual operation, the method in the first embodiment may improve the delivery effect of some advertisements, but have a negative impact on the delivery effect of other advertisements, this embodiment, based on the first embodiment, further allocates the playback order of each advertisement when the number of advertisements to be played and the required playback volume of each advertisement are determined. The specific steps are as follows:

[0090] According to the method in the second step of the first embodiment, the advertising playing benefit score R of each advertisement to be played is calculated in sequence;

[0091] Get the completion ratio cj of each advertisement to be played in the current cycle; where j ranges from 1 to m, and m represents the number of advertisements to be played;

[0092] The period may be one day, one week, one month, etc.;

[0093] The completion ratio refers to the sum of the advertising play benefit scores R corresponding to each play of the advertisement to be played in the current cycle / the sum of the advertising play benefit scores R required to be completed in the current cycle;

[0094] Calculate the completion ratio cj of each pending ad before a new pending ad is played, and then calculate the standard deviation S1 of these m cj values;

[0095] Calculate the standard deviation S2 of the advertisements to be played after the completion ratio cj is updated after playing each advertisement to be played;

[0096] Calculate m corresponding |S1-S2| values ​​(different updated values ​​correspond to different ads to be played), select the ad to be played corresponding to the S2 value with the smallest |S1-S2| value, and use the ad to be played as the next ad to be played by the display unit.

[0097] This application confirms the next advertisement to be played through the completion ratio of each advertisement and the real-time updated advertisement playback efficiency score, so as to ensure that the completion ratio progress of each advertisement to be played is similar, so as to ensure as much as possible that the advertisement to be played with a larger sum of the advertisement playback efficiency score R can be played when the corresponding single advertisement playback efficiency score R is larger, thereby improving the utilization efficiency of human flow resources while ensuring the playback requirements of all advertisements.

[0098] Example 3

[0099] Based on the second embodiment, this embodiment further proposes a method for adjusting the number of advertisements to be played on each display unit, as follows:

[0100] According to the method in the second embodiment, when a display unit completes playing of an advertisement to be played within a cycle;

[0101] First, check whether there is time left in the cycle. If so, add the number of ads to be played in the display unit.

[0102] If not:

[0103] For an advertisement, obtain the order g of its corresponding advertisement playing benefits R when it is sorted by size each time it is used as an advertisement to be played;

[0104] Get the average value G of the order g corresponding to each playback of the advertisement in the entire cycle;

[0105] Obtain an average value Gp of G values ​​of all advertisements to be played after a display unit completes playing of advertisements to be played within a cycle;

[0106] When the average value Gp is ​​greater than the preset value Gy, it indicates that the utilization efficiency of the corresponding display unit is low, and the number of advertisements to be played in the display unit can be increased.

[0107] On the contrary, if the average value Gp is ​​not greater than the preset value Gy, it means that the utilization efficiency of the corresponding display unit is high, and the number of advertisements to be played in the display unit is not adjusted.

[0108] This method can further increase the revenue generated by display unit advertising and improve the return on advertising.

[0109] Example 4

[0110] The present invention also discloses a system for executing the above-mentioned method for precise delivery of personalized advertising content based on user portraits, the system comprising:

[0111] An image acquisition unit is used to acquire scene images in the elevator and transmit the acquired scene image information to the image analysis unit;

[0112] An image analysis unit, configured to analyze scene images and identify the number, gender, and age of people therein;

[0113] The main control unit is used to determine the next advertisement to be played based on the portrait data of the people in the elevator analyzed by the image analysis unit;

[0114] Display unit, used for playing advertisements.

[0115] The above contents are merely examples and explanations of the present invention. Those skilled in the art may make various modifications or additions to the described specific embodiments or replace them in similar ways. As long as they do not deviate from the invention or exceed the scope defined by the claims, they should all fall within the scope of protection of the present invention.

Claims

1. A method for accurately delivering personalized advertising content based on user portraits, characterized in that: The steps include: The first step is to confirm the target audience of each advertisement and the corresponding profile data; Get the proportion of each audience type corresponding to each advertisement; The value of i ranges from 1 to n, where n represents the number of audience types; The second step is to collect image information inside the elevator through the image acquisition unit in the elevator at time t1 before the next advertisement is played; Where t1 is the preset value; Analyze and obtain the gender and age of each person in the elevator; Divide the people in the elevator into several character groups based on their gender and age, and each character group corresponds to a target audience type; Obtain each advertisement that can be played by the display unit and record it as an advertisement to be played; Obtaining the advertisement playing benefit score R corresponding to the composition of people in the current elevator when assuming that each advertisement to be played is played; In the third step, the advertisement to be played with the largest corresponding advertisement playing benefit score R is selected as the advertisement to be played next.

2. The method for accurately delivering personalized advertising content based on user portraits according to claim 1, characterized in that: When determining the number of audience types, sort the proportions of each audience group in descending order, and then add them up in descending order until the combined proportion exceeds the preset proportional coefficient. The audience groups involved in the summation calculation will be used as the final confirmed audience types for the advertisement, and the proportions will be normalized.

3. The method for accurate delivery of personalized advertising content based on user portraits according to claim 1, characterized in that: The calculation method of advertising play benefit score R is: Among them, ai represents the corresponding proportion of each character group, and bi represents the number of people in each character group in the elevator.

4. The method for accurate delivery of personalized advertising content based on user portraits according to claim 1, characterized in that: It also includes an advertising plan optimization strategy, including the following steps: Obtaining the advertising play benefit score R of each advertisement played by each display unit within a preset time range in the past; For a display unit, sort the advertisements played by it in descending order according to the sum of the corresponding advertisement play benefit scores R, which is expressed as ek; Where k represents the order of each advertisement, and ek represents the sum of the specific advertisement broadcast benefit scores R; Sort the ads by importance, and then determine the display units for each ad in that order to form several sets of plans. For one of the solutions, after all advertisements have completed the determination of display units, the order of each advertisement in the corresponding display units is calculated, and then the sum H of the orders corresponding to each advertisement is calculated; Calculate the sum H1 of the sum H of the corresponding orders of all advertisements; The H1 value corresponding to each plan is calculated in turn, and the plan with the smallest H1 is determined as the final advertising broadcast plan, and the advertisement is broadcast according to the advertising broadcast plan in the next cycle.

5. The method for accurate delivery of personalized advertising content based on user portraits according to claim 4, characterized in that: When the number of ads to be played changes: If the number of advertisements is reduced on the existing basis, delete the corresponding advertisements in the advertisement playing plan; When the number increases, the added advertisements are trial-played in each display unit and then the advertisement play plan optimization strategy is executed based on the play results.

6. The method for accurate delivery of personalized advertising content based on user portraits according to claim 4, characterized in that: When the number of ads to be played and the required playback volume of each ad are determined, the playback order of each ad is further allocated. The specific steps are as follows: Calculate the advertising benefit score R of each advertisement to be played in sequence; Get the completion ratio cj of each advertisement to be played in the current cycle; j ranges from 1 to m, where m represents the number of advertisements to be played; The completion ratio refers to the sum of the advertising play benefit scores R corresponding to each play of the advertisement to be played in the current cycle / the sum of the advertising play benefit scores R required to be completed in the current cycle; Calculate the completion ratio cj of each pending ad before a new pending ad is played, and then calculate the standard deviation S1 of these m cj values; Calculate the standard deviation S2 of the advertisements to be played after the completion ratio cj is updated after playing each advertisement to be played; Calculate m corresponding |S1-S2| values, select the advertisement to be played corresponding to the S2 value with the smallest |S1-S2| value, and use the advertisement to be played as the next advertisement to be played by the display unit.

7. The method for accurate delivery of personalized advertising content based on user portraits according to claim 6, characterized in that: The method for adjusting the number of advertisements to be played on each display unit is also included, as follows: When a display unit completes playing of advertisements to be played within a cycle; First, check whether there is time left in the cycle. If so, add the number of ads to be played in the display unit. If not, for an advertisement, obtain the order g of its corresponding advertisement playing benefit R when it is sorted by size each time it is used as an advertisement to be played; Get the average value G of the order g corresponding to each playback of the advertisement in the entire cycle; Obtain an average value Gp of G values ​​of all advertisements to be played after a display unit completes playing of advertisements to be played within a cycle; When the average value Gp is ​​greater than the preset value Gy, the number of advertisements to be played is increased in the display unit.

8. A personalized advertising content precision delivery system based on user portraits, characterized by: The system is used to execute the delivery method according to any one of claims 1 to 7, and the system includes: An image acquisition unit is used to acquire scene images in the elevator and transmit the acquired scene image information to the image analysis unit; An image analysis unit, configured to analyze scene images and identify the number, gender, and age of people therein; The main control unit is used to determine the next advertisement to be played based on the portrait data of the people in the elevator analyzed by the image analysis unit; Display unit, used for playing advertisements.

Images