Artificial intelligence-based display content self-adaptive optimization method and system

Abstract

The present application relates to the technical field of display content optimization, in particular to a display content adaptive optimization method and system based on artificial intelligence, the method comprising the following steps: acquiring scene information of a position where a target display is located and an advertisement video library to be played using the target display; setting a cloud database for the target display, dividing the cloud database into a main play chain, a feature information library and a plurality of secondary play chains; formulating display sorting rules for the main play chain; formulating trigger conditions for optimized play information for the advertisement video currently displayed by the target display; sorting a plurality of advertisement videos according to the display sorting rules to obtain a display order, and displaying the plurality of advertisement videos in sequence according to the display order, while triggering the operation of optimized play information according to the trigger conditions, and optimizing the play information of the currently displayed advertisement video, which can quickly adapt to changes in the scene where the display is located, and improve the flexibility and effectiveness of the displayed advertisement content.

Description

Artificial intelligence-based adaptive optimization method and system for display content Technical Field

[0001] This invention relates to the field of display content optimization technology, specifically to a display content adaptive optimization method and system based on artificial intelligence. Background Technology

[0002] In today's era of rapid development of digital information, displays, as key terminal devices for information display, are widely used in various fields, such as commercial advertising. Traditional advertising methods often use fixed playlists and sequences, which cannot dynamically adjust the display order of advertising videos according to real-time changing display scenarios and target audience characteristics, thus affecting the display effect of the content.

[0003] To address these issues, we propose an AI-based adaptive optimization method and system for display content to solve the aforementioned problems. Summary of the Invention

[0004] The purpose of this invention is to provide an artificial intelligence-based method and system for adaptive optimization of display content, in order to solve the problems mentioned in the background art.

[0005] To achieve the above objectives, the present invention provides the following technical solution: a display content adaptive optimization method and system based on artificial intelligence, the method comprising the following steps:

[0006] Obtain scene information of the target display's location and the advertising video library to be played on the target display; set up a cloud database for the target display and store the scene information and advertising video library in the cloud database;

[0007] The cloud database is divided into a main playback chain, a feature information database, and multiple secondary playback chains; display sorting rules are defined for the main playback chain; and triggering conditions for optimizing playback information are defined for the advertising video currently displayed on the target monitor.

[0008] The multiple ad videos are sorted according to the display sorting rules to obtain the display order. The multiple ad videos are then displayed sequentially according to the display order. At the same time, the playback information optimization operation is triggered according to the trigger conditions to optimize the playback information of the currently displayed ad video.

[0009] Preferably, the step of setting up a cloud database for the target display and storing the scene information and advertising video library in the cloud database includes:

[0010] The cloud database is divided into a main playback chain, a feature information database, and multiple secondary playback chains;

[0011] Extract multiple scene feature information from scene information, including feature category information and feature weight information;

[0012] Multiple sequentially arranged information points are set for the main playback chain, and each information point is used to store advertising videos with different feature categories.

[0013] Multiple sequentially arranged feature points are set for the feature information database, and these feature points are used to store different feature category information.

[0014] Multiple scene feature information is bound one-to-one with multiple secondary playback chains, where each secondary playback chain includes multiple advertising videos ordered in ascending order of playback duration.

[0015] Preferably, the step of formulating display sorting rules for the main playback chain includes:

[0016] Obtain the feature points and information points corresponding to multiple feature categories, and set up a synchronization chain between the feature points and information points corresponding to the same feature category.

[0017] The feature weight information of the feature points is acquired in real time and the feature points are sorted in descending order in real time. The information points are sorted in real time by following the feature points through the synchronization chain as the sorting rule.

[0018] Preferably, the step of setting the trigger conditions for optimizing playback information for the advertising video currently displayed on the target display includes:

[0019] The feature weight information corresponding to the advertisement video currently displayed on the target display is obtained as the first weight information, and the feature weight information with the largest value in the current main playback chain is obtained as the second weight information.

[0020] Obtain the difference information between the first weight information and the second weight information, wherein the difference information includes the weight difference and the corresponding rate of change;

[0021] The weight difference exceeding the preset condition is used as the trigger condition for optimizing playback information. The playback information of the currently displayed advertisement video is adjusted according to the difference information, which includes playback speed and required playback duration.

[0022] Preferably, the step of adjusting the playback information of the currently displayed advertisement video based on the difference information, wherein the playback information includes playback speed and required playback duration, includes:

[0023] Calculate the weight difference between the first weight information and the second weight information, and determine whether the weight difference exceeds the preset conditions;

[0024] When the weight difference exceeds the preset condition, the available duration of the advertisement video corresponding to the first weight information is estimated based on the rate of change of the weight difference as the required playback duration, and the advertisement video that meets the required playback duration is extracted from the secondary playback chain for playback.

[0025] When the weight difference does not exceed the preset condition, the playback speed of the advertisement video will be adaptively adjusted according to the available time.

[0026] Preferably, the step of sorting multiple advertising videos according to display sorting rules to obtain the display order includes:

[0027] Based on the display sorting rules, obtain the display order of information points in the main playback chain after real-time sorting;

[0028] Based on the real-time sorting and display order of information points, the advertising videos are selected sequentially from the secondary playback chains corresponding to the information points.

[0029] The selected ad videos are arranged according to the selection order to obtain the display order of multiple ad videos.

[0030] Preferably, the step of sequentially displaying multiple advertising videos according to the display order, and simultaneously triggering the operation of optimizing playback information according to triggering conditions to optimize the playback information of the currently displayed advertising video includes:

[0031] Obtain the display order of the ad videos in the main playback chain, and display the multiple ad videos on the target display in the order they appear.

[0032] When a weight difference in the display order meets the triggering condition, the playback information of the currently displayed advertisement video is optimized.

[0033] An AI-based adaptive optimization system for display content, applied to any of the AI-based adaptive optimization methods for display content described above, includes:

[0034] The storage module is used to obtain scene information of the target display's location and the advertising video library to be played on the target display; it sets up a cloud database for the target display and stores the scene information and advertising video library in the cloud database.

[0035] The settings module is used to divide the cloud database into a main playback chain, a feature information database, and multiple secondary playback chains; to define display sorting rules for the main playback chain; and to define trigger conditions for optimizing playback information for the currently displayed advertisement video on the target display.

[0036] The optimization module is used to sort multiple ad videos according to display sorting rules to obtain the display order, and then display the multiple ad videos sequentially according to the display order. At the same time, it triggers the operation of optimizing playback information according to the triggering conditions to optimize the playback information of the currently displayed ad video.

[0037] Compared with the prior art, the beneficial effects of the present invention are:

[0038] By sorting information points on the main playback chain in real time based on display sorting rules, and selecting advertising videos from the secondary playback chain according to the sorted information points, the display order of advertising videos is dynamically adjusted according to the real-time changing delivery scenarios and target audience characteristics. The display of the current advertising playback information is optimized according to real-time scene changes, realizing dynamic and flexible sorting of advertising videos and real-time optimization of playback information, improving the accuracy and effectiveness of advertising delivery, and enabling advertisements to reach the target audience more effectively. Attached Figure Description

[0039] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0040] Figure 1 is a schematic diagram of the method flow of the present invention;

[0041] Figure 2 is a schematic diagram of the cloud database of the present invention;

[0042] Figure 3 is a system structure block diagram of the present invention. Detailed Implementation

[0043] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0044] Example

[0045] Please refer to Figures 1 to 3. This invention provides a display content adaptive optimization method and system technical solution based on artificial intelligence: the display content adaptive optimization method and system based on artificial intelligence includes the following steps:

[0046] S1: Obtain scene information of the target display location and the advertising video library to be played on the target display; set up a cloud database for the target display and store the scene information and advertising video library in the cloud database;

[0047] S2: Divide the cloud database into a main playback chain, a feature information database, and multiple secondary playback chains; define display sorting rules for the main playback chain; define triggering conditions for optimizing playback information for the currently displayed advertisement video on the target display.

[0048] The steps for setting up a cloud database for the target display and storing scene information and advertising video library in the cloud database include: dividing the cloud database into a main playback chain, a feature information library, and multiple secondary playback chains; extracting multiple scene feature information from the scene information, wherein the scene feature information includes feature category information and feature weight information; setting multiple sequentially arranged information points for the main playback chain, each information point being used to store advertising videos corresponding to different feature categories; setting multiple sequentially arranged feature points for the feature information library, wherein the feature points are used to store different feature category information; and binding the multiple scene feature information to multiple secondary playback chains one-to-one, wherein the secondary playback chains include multiple advertising videos ordered sequentially according to the ascending order of playback duration.

[0049] It should be noted that dividing the cloud database into a main playback chain, a feature information database, and multiple secondary playback chains refers to dividing the cloud database into regions, resulting in multiple sub-databases. The main playback chain, feature information database, and secondary playback chains correspond to these sub-databases. The secondary playback chains store multiple advertising videos corresponding to different feature categories, essentially storing an advertising video library for each feature category. The sub-database corresponding to the main playback chain stores multiple advertising videos extracted from the multiple secondary playback chains, representing different feature categories within the current scene information. The feature information database is a sub-database storing multiple feature categories. These sub-databases are connected according to the connection relationships between the main playback chain, feature information database, and secondary playback chains. Each sub-database is divided into sub-database blocks, arranged in a linear fashion. The multiple sub-database blocks in the main playback chain are equivalent to multiple information points, each storing advertising videos from different feature categories. The advertising videos to be played in the current scene are aggregated based on the feature category information of the current scene to form the main playback chain. Each information point corresponds to an advertising video for a specific feature category. When the feature weight information of that feature category changes... When updating, the position of the feature category information needs to be adjusted, which in turn adjusts the position of the information points of the advertisement videos bound to the feature category information. This, in turn, updates the sorting of the main playback chain, allowing it to adjust the display order according to real-time scene information. Multiple sub-database blocks in the secondary playback chain store the advertisement videos to be played corresponding to the feature category information. Each sub-database of the secondary playback chain is associated one-to-one with the feature points in the feature information database. Multiple sub-database blocks in the feature information database, each corresponding to a feature point, are used to store different feature category information in the scene where the target display is located. Each feature category information corresponds to a feature point stored in the corresponding sub-database block. The position order of each feature point is automatically adjusted according to the descending order of the feature weight information, thus completing the automatic sorting of feature points. A synchronization chain is set between feature points and information points, which is equivalent to associating and binding the sub-database blocks in the feature information library with the sub-database blocks in the main playback chain. When the sorting of feature category information in the feature information library changes, it can drive the sorting of advertising videos on the information points in the main playback chain to change. The main playback chain adjusts the display order in real time according to the scene information, thereby improving the display effect of the content displayed on the monitor.

[0050] Specifically, the main playback chain coordinates the overall playback order of ad videos, the feature information database stores scene feature information, and the secondary playback chain stores sets of ad videos corresponding to different scene feature information. Scene feature information, including feature category information and feature weight information, is extracted from the scene information. Ad videos corresponding to each feature category are extracted from the ad video database and sorted in ascending order of ad video duration to generate multiple secondary playback chains. One ad video is extracted from each of the multiple secondary playback chains corresponding to each feature category in the scene information to obtain multiple ad videos to be played. These multiple ad videos are then sorted according to feature weight information to generate the main playback chain. When the cloud database is divided into the main playback chain, feature information database, and multiple secondary playback chains, the main playback chain has a dynamic adjustment function, which can update the sorting of information points and feature points in real time according to actual playback effects and scene changes. The feature information database adopts an efficient data storage structure for rapid retrieval and updating of scene feature information. Ad videos in the secondary playback chains are subdivided and stored according to different classification criteria to improve the selection effect of ad videos.

[0051] The steps for formulating display sorting rules for the main playback chain include: obtaining feature points and information points corresponding to multiple feature category information respectively; setting up a synchronization chain between feature points and information points corresponding to the same feature category information; obtaining feature weight information on feature points in real time and sorting feature points in descending order in real time; and sorting information points in real time by following feature points through the synchronization chain as the sorting rule.

[0052] It should be noted that when setting up a synchronization chain between feature points and information points corresponding to the same feature category information, the synchronization chain adopts data association or pointer reference to ensure the synchronous update of feature points and corresponding information points.

[0053] The steps for setting trigger conditions for optimizing playback information of the advertisement video currently displayed on the target display include: obtaining the feature weight information corresponding to the advertisement video currently displayed on the target display as the first weight information, obtaining the feature weight information with the largest value in the current main playback chain as the second weight information; obtaining the difference information between the first weight information and the second weight information, wherein the difference information includes the weight difference and the corresponding rate of change; using the weight difference exceeding a preset condition as the trigger condition for optimizing playback information, and adjusting the playback information of the currently displayed advertisement video according to the difference information, wherein the playback information includes the playback speed and the required playback duration;

[0054] It should be noted that the feature weight information with the largest value in the current main playback chain is used as the second weight information. Since the positions are adjusted in real-time and sorted in descending order, during this real-time adjustment process, if the value of the first weight information is less than the largest feature weight information, the position of the information point corresponding to the ad video with the first weight information needs to be adjusted to conform to the sorting rules. The largest feature weight information is selected from the weight information corresponding to each ad video at all information points. The first weight information is the feature weight information corresponding to the currently playing ad video. When the feature weight information corresponding to the currently playing ad video is less than or equal to the largest feature weight information corresponding to an unplayed ad video (i.e., the difference is small), the second weight information is selected. When the difference between the first and second weight information is zero, the playback speed of the ad video corresponding to the first weight information needs to be increased. At the same time, it is adjusted to the corresponding information position according to the sorting rules. When the first weight information is greater than the second weight information, that is, when the difference is greater than zero, the remaining available time is estimated based on the rate of change of the difference. The estimated remaining available time is the required playback time of the ad video in this playback chain. The playback speed is adjusted according to the remaining time. When the difference between the first and second weight information is positive, the current ad video is played normally. When the playback is completed, the ad video corresponding to the required playback time is selected again. In this way, the display content of the target display can be better optimized according to the distribution of people in the scene information, thereby improving the display effect of the ad video.

[0055] Specifically, the main playback chain contains multiple information points arranged in descending order based on feature weight information. The ad video at each information point is bound to the feature weight information corresponding to the feature category information. The multiple information points on the main playback chain change in real time according to the changes in multiple feature weight information. The feature weight information is determined based on the proportion of each feature category information in the scene information. The feature category information is allocated based on the characters in the scene information. For example, if there are multiple characters in the scene information, they can be divided according to their age to obtain characters of different age groups. Each age group can be used as a feature category information. Alternatively, it can be combined with the characters' gender to divide them, such as males and females of different age groups, which can also yield multiple different feature categories information. The ad videos in the ad video library are classified according to the feature category information. The multiple ad videos corresponding to each feature category information are arranged in ascending order of video duration to generate a secondary playback chain, which facilitates the selection of appropriate ad videos for playback based on the difference in value.

[0056] The feature weights of each feature category are updated in real time based on changes in scene information. This real-time ranking of information points according to their weight percentages drives the ranking of corresponding information points, ultimately leading to the main playback chain. The ads on these information points are then displayed sequentially. During playback, if the information point of the currently playing ad changes, the playback speed is adjusted. If the information point remains unchanged but its weight decreases relative to subsequent weights, the remaining duration is estimated based on the rate of change of the feature weight values. The corresponding ad video duration is then selected from the secondary playback chain. This dynamic adjustment of the information point order within the main playback chain, based on real-time changes in scene information, improves the matching degree between the played ads and the characters in the scene. This allows for rapid adaptation to changing user needs in different scenarios, enabling the monitor to display ad content that better suits the current scene and user requirements, enhancing ad targeting, and improving the quality and effect of the displayed content.

[0057] Adjusting the playback information of the currently displayed advertisement video based on the difference information, wherein the playback information includes the steps of playback speed and required playback duration, includes: calculating the weight difference between the first weight information and the second weight information, and determining whether the weight difference exceeds a preset condition; when the weight difference exceeds the preset condition, estimating the available duration of the advertisement video corresponding to the first weight information based on the rate of change of the weight difference as the required playback duration, and extracting the advertisement video that meets the required playback duration from the secondary playback chain for playback; when the weight difference does not exceed the preset condition, adaptively adjusting the playback speed of the advertisement video based on the available duration.

[0058] Specifically, adaptively adjusting the playback speed of ad videos based on available time means automatically adjusting the playback speed according to the available time. When the available time is less than the time required for the normal playback speed of the ad video but less than the playback time of the ad video with the shortest required playback time in the next playback chain, the playback speed is increased. The specific required playback speed is then adjusted based on the difference between the available time and the playback time. For example, if the feature weight information corresponding to the currently displayed ad video has the highest value, before it falls below other maximum feature weight information, and the available time is 120 seconds, and the currently displayed ad video needs 30 seconds to complete playback, then the remaining available time is 90 seconds. The system then finds the ad video in the next playback chain that is greater than and closest to 90 seconds. The advertisement video serves as the next advertisement video for the information point corresponding to the feature weight information and continues to play. Since the available time is less than the normal playback time of the advertisement video, the playback speed of the advertisement video needs to be adjusted. The playback speed is automatically adjusted based on the difference between the available time and the normal playback time. The larger the difference, the slower the playback speed, and the smaller the difference, the faster the playback speed. This allows for timely adjustment of the playback strategy based on changes in the feature weight information in the scene information, enabling a smoother transition to the advertisement video corresponding to the next feature category information. In turn, based on changes in the weight of the feature category information, the display content of the target display is modified in a timely manner, and the advertisement video is more accurately placed on the corresponding scene features, improving the display effect of the target display.

[0059] S3: Sort multiple ad videos according to the display sorting rules to obtain the display order, and display the multiple ad videos in sequence according to the display order. At the same time, trigger the operation of optimizing playback information according to the triggering conditions to optimize the playback information of the currently displayed ad video.

[0060] The steps for sorting multiple ad videos according to display sorting rules to obtain the display order include: obtaining the display order of information points in the main playback chain after real-time sorting based on the display sorting rules; selecting ad videos sequentially from the secondary playback chains corresponding to the information points according to the real-time sorted display order of the information points; and arranging the selected ad videos according to the selection order to obtain the display order of multiple ad videos.

[0061] Multiple ad videos are displayed sequentially according to their display order. At the same time, the operation of optimizing playback information is triggered according to the triggering conditions. The steps of optimizing the playback information of the currently displayed ad video include: obtaining the display order of ad videos in the main playback chain, displaying multiple ad videos on the target display in the order of display; and optimizing the playback information of the currently displayed ad video when there is a weight difference in the display order that meets the triggering conditions.

[0062] By sorting information points on the main playback chain in real time based on display sorting rules, and selecting advertising videos from the secondary playback chain according to the sorted information points, the display order of advertising videos is dynamically adjusted according to the real-time changing delivery scenarios and target audience characteristics. The display of the current advertising playback information is optimized according to real-time scene changes, realizing dynamic and flexible sorting of advertising videos and real-time optimization of playback information, improving the accuracy and effectiveness of advertising delivery, and enabling advertisements to reach the target audience more effectively.

[0063] An AI-based adaptive optimization system for display content, applied to any of the AI-based adaptive optimization methods for display content described above, includes:

[0064] The storage module is used to obtain scene information of the target display's location and the advertising video library to be played on the target display; it sets up a cloud database for the target display and stores the scene information and advertising video library in the cloud database.

[0065] The settings module is used to divide the cloud database into a main playback chain, a feature information database, and multiple secondary playback chains; to define display sorting rules for the main playback chain; and to define trigger conditions for optimizing playback information for the currently displayed advertisement video on the target display.

[0066] The optimization module is used to sort multiple ad videos according to display sorting rules to obtain the display order, and then display the multiple ad videos sequentially according to the display order. At the same time, it triggers the operation of optimizing playback information according to the triggering conditions to optimize the playback information of the currently displayed ad video.

[0067] By sorting the main playback chain information points in real time and selecting ad videos from the secondary playback chain, the display order of ad videos can be flexibly adjusted. The display information can be optimized in a timely manner according to the trigger conditions, which can better adapt to different advertising scenarios and target audiences and improve the playback effect of ad videos.

[0068] In the description of this specification, references to terms such as "an embodiment," "example," "specific example," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0069] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. An artificial intelligence-based adaptive optimization method for display content, characterized in that, Includes the following steps: Obtain scene information about the location of the target display and the library of advertising videos to be played on the target display; Set up a cloud database for the target display and store scene information and advertising video library in the cloud database; The steps for setting up a cloud database for the target display and storing scene information and an advertising video library in the cloud database include: dividing the cloud database into a main playback chain, a feature information library, and multiple secondary playback chains; extracting multiple scene feature information from the scene information, wherein the scene feature information includes feature category information and feature weight information; setting multiple sequentially arranged information points for the main playback chain, each information point being used to store advertising videos corresponding to different feature categories; setting multiple sequentially arranged feature points for the feature information library, wherein the feature points are used to store different feature category information; binding the multiple scene feature information to multiple secondary playback chains one-to-one, wherein the secondary playback chains include multiple advertising videos ordered in ascending order of playback duration; dividing the cloud database into the main playback chain, the feature information library, and multiple secondary playback chains; and further dividing the cloud database into the main playback chain, the feature information library, and multiple secondary playback chains. The steps for defining the display sorting rules for the main playback chain include: obtaining feature points and information points corresponding to multiple feature categories, setting synchronization chains between feature points and information points corresponding to the same feature category; obtaining feature weight information on feature points in real time and sorting feature points in descending order, and using the synchronization chains to follow feature points in real time as sorting rules; sorting multiple ad videos according to the display sorting rules to obtain the display order, displaying multiple ad videos sequentially according to the display order, and simultaneously triggering the operation to optimize the playback information of the currently displayed ad video according to the triggering conditions.

2. The display content adaptive optimization method based on artificial intelligence according to claim 1, characterized in that: The step of setting trigger conditions for optimizing playback information of the advertisement video currently displayed on the target display includes: obtaining feature weight information corresponding to the advertisement video currently displayed on the target display as first weight information, obtaining feature weight information with the largest value in the current main playback chain as second weight information; obtaining the difference information between the first weight information and the second weight information, wherein the difference information includes the weight difference and the corresponding rate of change; using the weight difference exceeding a preset condition as the trigger condition for optimizing playback information, and adjusting the playback information of the currently displayed advertisement video according to the difference information, wherein the playback information includes playback speed and required playback duration.

3. The display content adaptive optimization method based on artificial intelligence according to claim 2, characterized in that: The step of adjusting the playback information of the currently displayed advertisement video based on the difference information, wherein the playback information includes playback speed and required playback duration, includes: calculating the weight difference between the first weight information and the second weight information, and determining whether the weight difference exceeds a preset condition; when the weight difference exceeds the preset condition, estimating the available duration of the advertisement video corresponding to the first weight information based on the rate of change of the weight difference as the required playback duration, and extracting advertisement videos that meet the required playback duration from the secondary playback chain for playback; when the weight difference does not exceed the preset condition, adaptively adjusting the playback speed of the advertisement video based on the available duration.

4. The display content adaptive optimization method based on artificial intelligence according to claim 1, characterized in that: The step of sorting multiple advertising videos according to the display sorting rules to obtain the display order includes: obtaining the display order of information points in the main playback chain after real-time sorting based on the display sorting rules; selecting advertising videos sequentially from the secondary playback chains corresponding to the information points according to the real-time sorted display order of the information points; and arranging the selected advertising videos according to the selection order to obtain the display order of multiple advertising videos.

5. The display content adaptive optimization method based on artificial intelligence according to claim 1, characterized in that: The steps of sequentially displaying multiple advertising videos according to the display order and simultaneously optimizing the playback information based on triggering conditions include: obtaining the display order of advertising videos in the main playback chain, displaying multiple advertising videos sequentially on the target display according to the display order; and optimizing the playback information of the currently displayed advertising video when there is a weight difference in the display order that meets the triggering conditions.

6. An AI-based adaptive optimization system for display content, applied to the AI-based adaptive optimization method for display content as described in any one of claims 1-5, characterized in that, include: The storage module is used to obtain scene information about the location of the target display and a library of advertising videos to be played on the target display. The system sets up a cloud database for the target display, storing scene information and an advertising video library there. A configuration module divides the cloud database into a main playback chain, a feature information library, and multiple secondary playback chains. Display sorting rules are defined for the main playback chain. Trigger conditions for optimizing playback information are set for the advertising videos currently displayed on the target display. An optimization module sorts multiple advertising videos according to the display sorting rules to obtain a display order, displays the videos sequentially according to the display order, and simultaneously triggers playback optimization operations based on the trigger conditions to optimize the playback information of the currently displayed advertising video.

Images