Exhibition scene dynamic interactive advertisement generation method and AI decision system

Abstract

The invention relates to the technical field of man-machine interaction and digital display processing, and discloses an exhibition scene dynamic interactive advertisement generation method, which comprises the following steps: S1, receiving advertisement content data uploaded by exhibitors, and constructing the content data into displayable advertisement data units, wherein each advertisement data unit comprises a display position attribute and a display weight attribute; s2, collecting behavior data of the user in the display space, wherein the behavior data comprises a touch position, a sight focus point or a gesture interaction point; and S3, on the basis of the spatial association relationship between the display position of the advertisement data unit and the user behavior data, constructing a behavior-driven potential energy field for representing user interest distribution. According to the method, the advertisement content layout adjustment is guided by constructing the potential energy field model driven by the user behaviors, and the dynamic association matching of the users is realized, so that the on-site users can always contact the personalized and high-relevancy advertisement content in the moving process, and the advertisement click rate and the conversion efficiency are remarkably improved.

Description

Technical Field

[0001] The present invention relates to the technical fields of human-computer interaction and digital display, and particularly to a method for generating dynamic interactive advertisements in exhibition scenarios and an AI decision-making system. Background Art

[0002] With the continuous upgrading of digital information dissemination means, the advertising industry is undergoing a deep transformation from traditional offline media to digital interactive media. Driven by new-generation information technologies such as 5G, AI, and AR / VR, the presentation mode of advertising content has gradually changed from linear playback, fixed content, and one-way dissemination to an intelligent display system featuring multi-modal fusion, behavior perception-driven, and real-time content adaptation.

[0003] Especially in scenarios with high interaction density such as large-scale exhibitions, brand release conferences, commercial shopping guide scenarios, enterprise exhibition halls, cultural and tourism guides, and smart business districts, exhibitors and content providers generally hope to enhance user attention, strengthen brand communication effects, extend the interactive stay time, and ultimately achieve synchronous improvement of information conversion rate and marketing conversion rate through more immersive, intelligent, and differentiated content display methods.

[0004] In such application environments, advertising display systems face a series of new challenges: First, the behavior paths and interest preferences of users show highly dynamic and personalized characteristics, and traditional static content push methods are difficult to meet the need for instantaneously matching user attention points; Second, the advertising content itself is becoming increasingly diverse, evolving from simple graphics and texts to dynamic images, videos, interactive components, and even 3D models or virtual characters. Existing systems are difficult to coordinate the scheduling and combined display of multiple types of resources; Then, restricted by the display space and terminal resources, the advertising system needs to dynamically adjust the display layout according to the importance of the content and user responses, prioritize the presentation of high-value information, and at the same time avoid visual obstacles such as content overlap and interface congestion; The association between display behavior and user feedback becomes the key to the data closed-loop. How to convert users' browsing, clicking, and interaction behaviors into the system's weight adjustment and display optimization mechanism is a technical difficulty that the current intelligent display system urgently needs to break through. Summary of the Invention

[0005] To make up for the above deficiencies, the present invention provides a method for generating dynamic interactive advertisements in exhibition scenarios and an AI decision-making system, aiming to improve the technical problems in existing advertising display systems such as the lack of user behavior perception, low matching degree between display content and user interests, crowded and chaotic spatial arrangement, and the lack of dynamic weight update and closed-loop optimization mechanism.

[0006] In a first aspect, the present invention provides the following technical solution. A method for generating dynamic interactive advertisements in exhibition scenarios includes: S1. Receive the advertisement content data uploaded by exhibitors, and construct the content data into displayable advertisement data units, where each advertisement data unit includes a display position attribute and a display weight attribute; S2. Collect the behavior data of users in the display space, where the behavior data includes touch positions, line-of-sight focus points, or gesture interaction points; S3. Based on the spatial association relationship between the display positions of the advertisement data units and the user behavior data, construct a behavior-driven potential field for characterizing the user interest distribution; S4. Dynamically adjust the display positions and display weights of the advertisement data units according to the change trend of the potential field; S5. Based on the updated state of the advertisement data units, generate personalized advertisement display content for a touch interface or an augmented reality space.

[0007] Preferably, the advertisement content data in S1 includes images, texts, videos, or three-dimensional models, and each advertisement data unit is associated with a unique exhibitor identification information for subsequent placement tracking.

[0008] Preferably, the behavior data in S2 is collected by a terminal device, and the terminal device includes a touch screen, an augmented reality headset, or an interactive terminal equipped with a line-of-sight tracking function.

[0009] Preferably, the steps for constructing the potential field in S3 include: constructing a corresponding interest attraction value according to the spatial distance between the advertisement data unit and the user behavior point and the current display weight, and performing weighted superposition on the display space to form a potential energy distribution.

[0010] Preferably, the process of dynamically adjusting the display weight of the advertisement data unit in S4 includes: increasing or decreasing the weight attribute according to the behavior feedback score of the user for the advertisement content and the exposure fatigue information, and the behavior feedback score is calculated based on click, stay, and interaction behaviors.

[0011] Preferably, the generation of the personalized advertisement display content in S5 is based on the positions and weight states of all current advertisement data units, and a content level scheduling strategy is adopted to display high-weight advertisement content in the main form and display low-weight content in the secondary form or hide it.

[0012] Preferably, the method further includes optimizing the balance of the global spatial distribution of the advertisement data units to make the potential field distribution in the display area more uniform, so as to avoid information overlap or concentration.

[0013] In a second aspect, the present invention provides the following technical solution, a dynamic interactive advertisement generation system for a convention and exhibition scenario, including: A data access module for receiving exhibitor advertising content data and user behavior data; An AI decision-making module for constructing a potential field model based on user behavior data and advertising content data, identifying the distribution of user interests, and judging the display priority and position adjustment strategy of the advertising content accordingly; A weight update module for combining user feedback data and display fatigue information to update the display weight of the advertising content in real time; A content generation module for generating personalized advertising content for the display terminal according to the updated content status combination; A display output module for presenting the generated advertising content on a touch screen or an augmented reality device.

[0014] Thirdly, the present invention provides the following technical solution. A computer device includes a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, it implements the above-mentioned dynamic interactive advertising generation method for a convention and exhibition scenario.

[0015] Fourthly, the present invention provides the following technical solution. A readable storage medium stores a computer program, and when the computer program is executed by a processor, it implements the above-mentioned dynamic interactive advertising generation method for a convention and exhibition scenario.

[0016] The present invention has the following beneficial effects: 1. In the present invention, by constructing a potential field model driven by user behavior to guide the adjustment of the advertising content layout, the dynamic correlation matching of the user's staying position, click behavior and the displayed advertising content is realized, and the effect that on-site users can always contact personalized and highly relevant advertising content during movement is obtained, significantly improving the advertising click-through rate and conversion efficiency.

[0017] 2. In the present invention, by introducing an attraction weight update mechanism based on user feedback and fatigue, the self-regulation and rotation control of the advertising display frequency and display content are realized, and the effect of continuously maintaining user attention and avoiding advertising fatigue in high-frequency exposure scenarios such as conventions and exhibitions is obtained, effectively improving the advertising exposure quality.

[0018] 3. In the present invention, by designing a potential energy balance optimization strategy for the display space, the intelligent distribution adjustment of the advertising content in the physical exhibition area or virtual interface is realized, and the effect that even in a high-density crowd or complex layout environment, the interface can be kept clean, the display is clear, and the content does not block each other is obtained, enhancing the user experience and display efficiency.

[0019] 4. In the present invention, by constructing a multi-modal terminal data fusion mechanism, the unified perception and processing of user behavior data by different types of interactive terminals are realized, and the system can be smoothly deployed and stably operated in various exhibition display device environments, greatly reducing the system adaptation and maintenance costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 It is a flowchart of a method for generating a dynamic interactive advertisement in an exhibition scenario proposed by the present invention; Figure 2 It is a system architecture diagram of a system for generating a dynamic interactive advertisement in an exhibition scenario proposed by the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0021] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Apparently, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0022] Embodiment 1 Referring to Figure 1 , in the first embodiment of the present invention, the present invention provides a method for generating a dynamic interactive advertisement in an exhibition scenario, including: S1. Receive the advertisement content data uploaded by the exhibitor, and construct the content data into a displayable advertisement data unit, where each advertisement data unit includes a display position attribute and a display weight attribute; the advertisement content data includes images, texts, videos or three-dimensional models, and each advertisement data unit is associated with a unique exhibitor identification information for subsequent delivery tracking.

[0023] Specifically, in this embodiment, the system first executes a data initialization process. The system receives the data uploaded by the exhibitor from the exhibition background platform. The data source can be in various forms: static images, product text descriptions, embedded video files, three-dimensional interactive models, etc. After parsing these materials, structured advertisement content data units are generated. This unit is the core data carrier, and all subsequent behavior analysis and layout calculations are carried out around this unit.

[0024] Each advertisement data unit internally contains the following core fields: Display position attribute (Position): Initially specified by the system or set by the exhibitor, usually represented by two-dimensional coordinates (x, y) or three-dimensional coordinates (x, y, z); Display weight attribute (Weight): Initially set to 1.0, indicating the basic attraction value. It will be dynamically adjusted according to user behavior feedback later; Data type (Type): including Image, Text, Video, Model; Content path or link (URL / Path): used to show loading; Unique Exhibitor Identifier (SID): used for content attribution, delivery tracking and feedback statistics.

[0025] In addition, the system generates a unique content identifier (CID) for each ad unit, which is combined with the SID to form a composite primary key to ensure the scalability of data tracking and analysis.

[0026] The data unit construction process is not a simple stacking of materials, but a structured process. For example, after a product image is uploaded, it is not directly used as display content, but is encapsulated as an AdUnit. , along with its display intention and recommended location. For example, the recommended location could be: "Within 5 meters of the entrance, near the right wall". This description will be converted into spatial coordinates and used as the initial location.

[0027] Display weight is not a static property. As an important factor in the optimization calculation of the AI decision-making system, its existence must be made clear in the initial construction stage. The subsequent update mechanism will be based on the following: ; Among them: Score_feedback is the user behavior feedback score, which combines the number of clicks, stay time, and interaction frequency; Fatigue represents display fatigue, which is calculated as follows: ;coefficient , β, γ, and δ can be adjusted based on the actual exhibition environment.

[0028] The constructed advertising data unit will be written into the cache pool and allocated to the current active scene for subsequent potential modeling and AI control module reading. Each data unit will undergo a "content score" before being called by the system, which is the first step of the AI decision-making system. The initial display priority is calculated based on the exhibitor level, the advertising content quality score (given by the background content reviewer), and the relevance label to the current scene. The priority will determine whether the advertising unit is displayed as the main position in the initial layout.

[0029] The AI scheduler in the system will also intervene at this stage to make preliminary classifications. For example, video ads are more likely to be placed in interactive areas, while graphic information may be placed on the side of the aisle or near information boards. This strategy is guided by rules and driven by historical data. It does not rely entirely on manual configuration and has self-learning capabilities.

[0030] The complete identification system (SID + CID) introduced in this stage can provide strong support for subsequent display tracking and the construction of exhibitor behavior portraits. For example, if the number of clicks on a certain advertising unit is much higher than that of other content, the system can trace back to the specific file of this content through CID, and then trace back to the exhibitor through SID to complete conversion rate analysis and value attribution.

[0031] More importantly, the processing in this stage provides a complete data basis for subsequent potential energy field modeling. Through the initialization of the position and weight of data units, the system can quickly construct the following potential energy function in the next potential energy modeling: ; where ( , ) is the current coordinate of the advertising unit, is its current display weight, is the smoothing factor to prevent division by zero. This potential energy function will be used subsequently to judge the attraction relationship between the user behavior interest points and the advertising distribution, so as to realize the intelligent evolution process of "content actively approaching people".

[0032] S2. Collect the behavior data of users in the display space. The behavior data includes touch positions, line-of-sight focus points, or gesture interaction points; the behavior data is collected through terminal devices, and the terminal devices include touch screens, augmented reality headsets, or interactive terminals equipped with line-of-sight tracking functions.

[0033] Specifically, in this embodiment, to achieve precise dynamic adaptation between the advertising content and the current user attention, the system needs to collect the behavior data of users in the display space in real time. The behavior data is the core basis for the system to judge the user interest positions, and its collection method and accuracy directly determine the accuracy of subsequent energy field modeling and advertising scheduling.

[0034] The sources of the behavior data include but are not limited to the following three categories: The first category is touch position data. It is applicable to display terminals equipped with multi-touch screens, such as interactive information screens, navigation panels, etc. When the user's finger touches any position on the screen, the system records the corresponding two-dimensional coordinates , and marks the current timestamp . The system adopts a frame synchronization acquisition mechanism to ensure that the behavior response delay is less than 50 milliseconds to avoid judgment errors.

[0035] The second category is line-of-sight focus point data. It is applicable to augmented reality headsets (ARHMDs) or terminal devices equipped with eye trackers (EyeTrackers). The system captures the user's fixation points in real time through ray tracing and perspective model fitting algorithms. The capture result is a three-dimensional coordinate point , where respectively represent the spatial positions of the fixation points in the exhibition space. To improve stability, the system adopts a fixation duration threshold mechanism. Only when the fixation duration exceeds the set time will it be considered that the user has generated a clear attention behavior.

[0036] The third type is gesture interaction point data. It is applicable to devices that support spatial gesture interaction, such as LeapMotion controllers or AR terminals with hand recognition functions. The system obtains the hand movement path in real time and recognizes key gestures, such as "click", "point", "grab". After recognition, the system extracts the corresponding spatial coordinates and binds the event type (such as Pointing for finger pointing and Tap for clicking).

[0037] All behavior points are added to the behavior cache queue and processed according to time windows. To improve the accuracy of behavior interest determination, the system adopts a weighted average strategy to calculate the user's "current behavior interest center". Let the behavior point sequence be where each point contains coordinates or , then the calculation method of the interest point center coordinates is as follows: For a two-dimensional scene: ; For a three-dimensional scene: ; Among them: : is the weight of each behavior point, initially assigned according to the behavior type. For example, Touch is 1.0, Gaze is 0.8, and Gesture is 0.9; : is the coordinate of the th behavior point; : is the total number of behavior points collected within the current window.

[0038] The system also has a behavior data filtering mechanism to exclude jitter, invalid signals, or high-frequency behavior without feedback in a short period. For example, short-term rapid saccades are not counted as valid fixation points; areas with continuous touches but no interaction feedback will be set as low-interest areas.

[0039] The acquisition frequency and processing rhythm of behavior data are controlled by a dynamic scheduling mechanism. When the user enters the exhibition area, the system runs at a high sampling frequency (e.g., 10Hz). When the user leaves or there is no interaction for a long time, the system switches to a low-power acquisition mode (1Hz or below) to save system resources.

[0040] S3. Based on the spatial association relationship between the display positions of advertising data units and user behavior data, construct a behavior-driven potential field for characterizing the user interest distribution; the steps for constructing the potential field include: constructing the corresponding interest attraction value according to the spatial distance between the advertising data unit and the user behavior point and the current display weight, and performing weighted superposition on the display space to form a potential energy distribution.

[0041] Specifically, in this embodiment, to achieve the spatial adaptability adjustment between the advertising content and the user interest area, the system constructs a "potential field model" based on user behavior data to simulate the spatial distribution trend of user interests in the current display space.

[0042] The potential field (PotentialField) is a calculation structure that draws on the physical gravitational model and is used to simulate the attraction effect of users on advertising units at different positions. In the system, each user behavior point will be regarded as an "interest source", and each advertising data unit will be regarded as a "force-receiving body". The stronger the user's behavioral interest and the closer the position is to a certain advertisement, the higher the probability that the advertisement content will be attracted and adjusted.

[0043] Two key inputs are required to construct this potential field model: the current position coordinates of each advertising data unit and the set of user behavior points within the current time window. The user interest points or behavior point sequences have been calculated according to the aforementioned S2.

[0044] First of all, the system needs to define each advertising unit For any user behavior point The interest attraction value . This value reflects the potential interest intensity of the user in the advertising unit in the current scenario and is defined as follows: ; Where: The user behavior point The attraction value of the advertising unit ; : The current display weight of the advertising unit (dynamic attribute, which can be updated by Al); Type definition, example: Touch = 1.0, Gaze = 0.8, Gesture = 0.9; : The advertising unit The Euclidean distance between and the behavior point is defined as follows: If it is a two-dimensional scenario: ; If it is a three-dimensional scenario: ; A small constant to prevent division by zero, usually taking .

[0045] The interest attraction value represents the attraction intensity of a single point to a single advertisement. However, the advertising system needs to construct an overall potential energy field distribution map to simulate the user interest trend in the entire space. The system expands the above attraction value into a spatial function, defined as the user potential energy field function or , as follows: Two-dimensional space: ; Three-dimensional space: ; Where: : The number of advertisement units in the current display space; : The number of valid user behavior points within the current time window; : The advertisement unit 's current position; : The coordinate of the behavior point ; : The interest potential value of the current spatial position point ; Through the above formula, the system can construct a continuous potential distribution map within the entire display space. The area with a higher potential value is more likely to be the current focus area of users and is suitable for focused layout of advertisement content.

[0046] S4. Dynamically adjust the display position and display weight of each advertisement data unit according to the change trend of the potential energy field; the process of dynamically adjusting the display weight of the advertisement data unit includes: adjusting the weight attribute by increasing or decreasing according to the behavior feedback score of the user for the advertisement content and the exposure fatigue information, and the behavior feedback score is calculated based on click, stay, and interaction behaviors.

[0047] Specifically, in this embodiment, to achieve continuous matching between advertisement content and user interests, the system needs to dynamically update the display weight of the advertisement data unit. This update is based on two core input sources: the user behavior feedback score and the display fatigue index. The advertisement display weight is not a fixed value but evolves continuously with interaction behaviors. The system uses a weight update mechanism to give higher display priorities to advertisements that users pay high attention to and give positive feedback, while weakening or even hiding content with weak feedback or causing fatigue.

[0048] First, define the user behavior feedback score. This score is calculated for each advertisement unit and combines the following multiple indicators: the number of user clicks , the total advertisement stay duration , the number of valid interactions ; The calculation formula of the feedback score is as follows ; Among them: Total number of clicks on the advertisement by the user within the set time window; Accumulated stay time, in seconds; : Maximum reference stay time set by the system to avoid abnormal scoring due to excessive time; : Number of effective interaction behaviors between the user and the advertisement (such as dragging, zooming, voice, etc.); : Weight coefficient, satisfying , used to control the relative importance of different behaviors.

[0049] Secondly, define the display fatigue index . This index measures the degree of "visual fatigue" caused by the frequent exposure of the advertisement content but lack of user response.

[0050] The fatigue is calculated as follows: ; Among them: : Advertisement unit Display fatigue of; : Total number of advertisement exposures (i.e., seen by the user but without clicks or interactions); : The same as in the feedback score; : Adjustment coefficient to control the balance between display and response; the higher the fatigue, the more difficult it is for the advertisement to trigger effective behaviors despite its frequent appearance, indicating that its display value is decaying.

[0051] The system calculates a new value of the advertisement display weight based on the feedback score and the fatigue . The update strategy takes the following form: ; Among them: : Updated advertisement display weight; : Display weight in the previous cycle; : Weight update learning rate, controlling the update amplitude, with typical values ranging from 0.1 to 0.3; : User feedback score; : Display fatigue; : Fatigue cancellation coefficient, used to adjust the degree of reduction of fatigue on the overall weight.

[0052] To prevent excessive weight oscillation, the system also sets upper and lower boundary limits: ; Among them: : Minimum allowable value of the display weight; : Maximum allowable value of the display weight; the conventional value is set to ; The weight update cycle can be triggered in two ways: Time-driven: The system automatically calculates once every fixed time (such as every 10 seconds); Event-driven: When a specific behavior event is detected, such as a click or gaze staying above a threshold, a local update is triggered immediately.

[0053] The real-time change of weight directly affects the display level of the content in the interface. High-weight ads will be promoted to the main display position by the system, or enlarged for presentation; ads with continuously decreasing weights will be gradually replaced or removed to free up resource positions.

[0054] S5. Based on the updated state of the advertising data unit, generate personalized advertising display content for the touch interface or augmented reality space. The generation of personalized advertising display content is based on the position and weight state of all current advertising data units, and adopts a content level scheduling strategy to display high-weight advertising content in the primary form and low-weight content in the secondary form or hide it.

[0055] It further includes optimizing the global spatial distribution of the advertising data units for balance, so that the potential energy field distribution in the display area is more uniform, thereby avoiding information overlap or concentration.

[0056] Specifically, this phase aims to map the updated advertising data units to personalized advertising content on a specific display interface. The display terminal can be a two-dimensional touch interface or a three-dimensional augmented reality (AR) space. The system dynamically builds content display strategies based on the current position status and latest weights of all advertising units to ensure personalized adaptation of advertising display in terms of form, position, and visual emphasis.

[0057] First, the system aggregates all current ad data units and reads the following key status parameters: ad unit number ; Display the current position coordinates in the space or -Current display weight , the dynamic calculation result from the previous stage; unit size parameters (width, height or bounding box size); content type label (graphics, video, interactive, etc.).

[0058] On this basis, the system implements the "content level scheduling strategy" for the advertising data units, dividing them into different display levels to achieve fine allocation and focus of display resources. The content level scheduling strategy is based on a multi-factor comprehensive scoring model. The system calculates a priority score for each advertising unit. , which determines its visual grade in the display space: ; in: : Current advertising weight; : The distance between the ad unit and the current user's interest center (Euclidean distance); : A small constant to prevent division by zero; : The potential field intensity at the current position; : The weight factor, satisfying , which can be adjusted according to the scenario.

[0059] According to After sorting the scores from high to low, the system divides all advertising content into three display levels: Level 1 content: The main display content. The score is in the top 10% or higher than the significant threshold, and usually adopts large-size exhibition booths, dynamic display forms, such as full-screen videos, floating 3D objects or key card displays.

[0060] Level 2 content: Secondary display content. The score is in the middle distribution range, and is displayed in lightweight forms such as thumbnails, small picture cards, static graphics and texts, and is often attached to the corners of the interface or the auxiliary information area.

[0061] Level 3 content: Low-priority display content. The score is relatively low, and the system can choose not to display it temporarily, or delay loading it as "candidate content" for alternative switching or scrolling display areas.

[0062] After the content level is divided, the system maps the advertising data unit to the actual display interface.

[0063] Embodiment 2: Referring to Figure 2 , in the second embodiment of the present invention, the present invention provides a dynamic interactive advertising generation system for exhibition scenarios, including: A data access module for receiving exhibitor advertising content data and user behavior data; An AI decision-making module for constructing a potential field model based on user behavior data and advertising content data, identifying the distribution of user interests, and judging the display priority and position adjustment strategy of advertising content accordingly; A weight update module for combining user feedback data and display fatigue information to update the display weight of advertising content in real time; A content generation module for generating personalized advertising content for the display terminal according to the updated content status combination; A display output module for presenting the generated advertising content on a touch screen or an augmented reality device.

[0064] Specifically, the overall architecture of the system follows the logical chain of "perception - analysis - decision - generation - presentation" to realize the real-time dynamic matching of advertising content and user interests.

[0065] The starting point of the system is the data access module, which is the input port of the system and connects the advertising content provider and the front end of the display terminal. This module continuously receives content data such as graphics, texts, and videos uploaded by advertisers and accesses the real-time behavior feedback of users, including events such as clicks, stays, slides, and interactions. All data is uniformly converted into a structured format and written into the behavior database or temporary buffer of the system for subsequent modules to call in real time.

[0066] The AI decision-making module is the intelligent center of the entire system, directly reading the user behavior and advertising content data provided by the data access module. Using this data as input, it establishes a user interest distribution model in the current display space to identify which areas are most concerned and which content is frequently ignored. The output of this module is a set of display priority scores and a spatial interest hotspot map, which will be used as the basis for content generation and weight update.

[0067] The weight update module runs in coordination with the AI decision-making module. Taking the user interaction intensity, display times, and feedback efficiency as the core inputs, it continuously updates the display weight of each advertisement. The updated weight data will be immediately fed back to the AI decision-making module for the next round of interest recognition and priority sorting. At the same time, the weight data is also synchronously transmitted to the content generation module to drive the hierarchical division of advertising content and the adjustment of display methods.

[0068] The content generation module receives the display score results from the AI decision-making module and the latest weight status from the weight update module. Based on this, it dynamically determines the display level (primary display, secondary display, candidate hidden) of each advertisement, the position arrangement in the display space, the display size, style, and presentation form, etc. If it detects uneven information density or crowded arrangement in the interface, this module will autonomously initiate the space optimization process to fine-tune the positions of advertisement units to ensure clear and orderly display.

[0069] Finally, all arrangement and display parameters are received by the display output module and converted into specific visual output signals. This module directly controls the display hardware at the front end and is responsible for rendering each advertisement content in real time according to the specified method. It also supports multi-terminal adaptation, such as flat screens, AR devices, or holographic projection interfaces, and can reload new display configurations according to subsequent feedback to achieve dynamic update and visual continuity of content.

[0070] The entire system forms a complete closed loop during operation. Each frame of content output by the display may trigger new user behaviors, which will be captured and transmitted back by the data access module, thus triggering the next round of iterative updates of AI analysis, weight adjustment, content reorganization, and re-rendering. This mechanism supports millisecond-level refreshing and second-level layout optimization, making the system highly responsive and adaptable in the long term.

[0071] Embodiment 3 Embodiment 3 of the present invention. Based on the same inventive concept, the present invention provides a computer-readable storage medium storing a computer program, and when the computer program is executed by a processor, the steps of a method for generating a dynamic interactive advertisement in an exhibition scenario according to the above embodiment are implemented.

[0072] Embodiment 4 Embodiment 4 of the present invention. Based on the same inventive concept, the present invention provides a terminal including: a processor and a memory; the processor and the memory communicate with each other; the memory is used for storing instructions; the processor is used for executing the instructions in the memory to implement a method for generating a dynamic interactive advertisement in an exhibition scenario according to the above embodiment.

[0073] It should be understood that each part of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, any one or a combination of the following well-known technologies in the art can be used: discrete logic circuits with logic gate circuits for implementing logical functions on data signals, application-specific integrated circuits with suitable combinational logic gate circuits, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0074] Finally, it should be noted that the above are only the preferred embodiments of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements on some of the technical features. Any modifications, equivalent replacements, improvements, 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 method for generating dynamic interactive advertisements in an exhibition and conference scenario, characterized in that, Including: S1. Receive the advertisement content data uploaded by exhibitors and construct the content data into displayable advertisement data units, where each advertisement data unit includes a display position attribute and a display weight attribute; S2. Collect the behavior data of users in the display space, and the behavior data includes touch positions, line-of-sight focus points or gesture interaction points; S3. Based on the spatial association relationship between the display positions of the advertisement data units and the user behavior data, construct a behavior-driven potential field for characterizing the user interest distribution; S4. Dynamically adjust the display positions and display weights of each advertisement data unit according to the change trend of the potential field; S5. Based on the updated state of the advertisement data units, generate personalized advertisement display content for the touch interface or the augmented reality space.

2. The dynamic interactive advertisement generation method for a convention and exhibition scene according to claim 1, wherein In S1, the advertisement content data includes images, texts, videos or three-dimensional models. Each advertisement data unit is associated with a unique exhibitor identification information for subsequent delivery tracking.

3. A method for generating a dynamic interactive advertisement in an exhibition scenario according to claim 1, characterized in that, In S2, the behavior data is collected through a terminal device, and the terminal device includes a touch screen, an augmented reality headset or an interactive terminal equipped with a line-of-sight tracking function.

4. A method for generating a dynamic interactive advertisement in an exhibition scenario according to claim 1, characterized in that, The steps for constructing the potential field in S3 include: constructing a corresponding interest attraction value according to the spatial distance between the advertisement data unit and the user behavior point and the current display weight, and performing weighted superposition on the display space to form a potential energy distribution.

5. A method for generating a dynamic interactive advertisement in an exhibition scenario according to claim 1, characterized in that, The process of dynamically adjusting the display weight of the advertisement data unit in S4 includes: increasing or decreasing the weight attribute according to the behavior feedback score of the user for the advertisement content and the exposure fatigue information, and the behavior feedback score is calculated based on click, stay and interaction behaviors.

6. The dynamic interactive advertisement generation method for a convention and exhibition scene according to claim 1, wherein The generation of the personalized advertisement display content in S5 is based on the positions and weight states of all current advertisement data units, and adopts a content level scheduling strategy to display the high-weight advertisement content in the main form, and display the low-weight content in the secondary form or hide it.

7. A method for generating a dynamic interactive advertisement in an exhibition scenario according to claim 1, characterized in that, The method further includes optimizing the global spatial distribution of the advertisement data units to make the potential field distribution in the display area more uniform, so as to avoid information overlap or concentration.

8. A dynamic interactive advertisement generation system for exhibition scenarios, characterized in that, A method for generating a dynamic interactive advertisement in an exhibition scenario according to any one of claims 1-7, including: A data access module for receiving exhibitor advertisement content data and user behavior data; An AI decision-making module for constructing a potential field model based on the user behavior data and the advertisement content data, identifying the user interest distribution, and judging the display priority and position adjustment strategy of the advertisement content accordingly; A weight update module for combining the user feedback data and the display fatigue information to update the display weight of the advertisement content in real time; A content generation module for generating personalized advertisement content for the display terminal according to the updated content state combination; A display output module for presenting the generated advertisement content on a touch screen or an augmented reality device.

9. A computer device, comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements a method for generating a dynamic interactive advertisement in an exhibition scenario according to any one of claims 1 to 7.

10. A readable storage medium, characterized in that, A computer program is stored on the readable storage medium, and when the computer program is executed by a processor, it implements a method for generating a dynamic interactive advertisement in an exhibition scenario according to any one of claims 1 to 7.

Images