Advertisement information flow management method and platform applied to multivariate visual design

Abstract

The invention discloses an advertisement information flow management method and platform applied to multivariate visual design, and belongs to the technical field of advertisement design, and the method specifically comprises the steps: dividing a screen interface into dynamic grid regions, and carrying out the comprehensive calculation based on user interaction behaviors and content semantic features, mapping the dynamic grid region into a semantic gravitational field representing user attention distribution; generating an initial advertisement container in a low-gravitational area at the edge of the screen, driving the container to migrate to a high-gravitational area according to the intensity gradient distribution of the semantic gravitational field, performing path optimization in the migration process, and dynamically adjusting the form of the container according to the visual density of a matched target area; dynamically adjusting the parameters of the advertisement container according to the semantic features of the native content of the target area and the real-time interaction behavior of the user; the cognitive load of the user is evaluated in real time according to the interaction behavior of the user on the surface of the advertisement container, information level folding is triggered when cognitive overload is detected, and secondary information or all information of the advertisement is dynamically hidden.

Description

Technical Field

[0001] The present invention relates to the technical field of advertising design, and in particular to an advertising information flow management method and platform applied to multi-dimensional visual design. Background Art

[0002] In today's digital age, advertising is ubiquitous across the diverse visual interfaces of various electronic devices, becoming a crucial tool for commercial promotion and information dissemination. With the rapid development of internet technology, users are exposed to a massive amount of advertising information daily. Applications and websites across the globe strive to efficiently display ads within limited screen space to attract user attention and generate commercial value. In this context, effectively managing the flow of advertising information to accurately reach target users while enhancing advertising effectiveness without compromising user experience has become a critical issue urgently needed in the advertising industry and related technology fields.

[0003] Currently, common approaches to managing ad information flow rely primarily on traditional ad placement strategies. For example, these methods employ simple interest matching based on user browsing history, search history, and other data, placing ads on user interfaces that may be of interest. While these approaches can achieve targeted ad placement to a certain extent, they lack the ability to accurately capture user attention and deeply integrate ad display with user interaction.

[0004] Existing ads often appear in a fixed form and position on the screen, and don't dynamically adjust based on changes in other onscreen content or user actions. This makes ads appear obtrusive and easily distracts users from the main content. Furthermore, traditional technologies struggle to accurately grasp the distribution of user attention in different scenarios and fail to fully integrate interface content to capture user attention. Consequently, ads lack targeted targeting, often failing to capture users' genuine attention and significantly reducing their effectiveness. Summary of the Invention

[0005] The purpose of the present invention is to provide an advertising information flow management method and platform for multi-dimensional visual design to solve the following technical problems: Existing advertisements often appear on the screen in relatively fixed forms and positions, and it is difficult to accurately grasp the distribution of users' attention in different scenarios, and they fail to fully integrate with the interface content to attract users' attention.

[0006] The purpose of the present invention can be achieved through the following technical solutions: An advertising information flow management method applied to multi-dimensional visual design includes the following steps: Real-time capture of user interaction behaviors, including touch and slide behavior data and interface dwell characteristics, while analyzing the semantic features of the current interface content, including text semantic attributes, image visual features, and video dynamic attributes; The screen interface is divided into dynamic grid areas. Based on the comprehensive calculation of user interaction behavior and content semantic features, the dynamic grid areas are mapped into a three-dimensional semantic gravity field that represents the distribution of user attention. The areas with higher user attention have higher gravity values. An initial ad container is generated in the low-gravity area at the edge of the screen. Based on the intensity gradient distribution of the semantic gravitational field, the container is driven to migrate to the high-gravity area. During the migration process, the path is optimized through fluid dynamics simulation, and the container shape is dynamically adjusted based on the visual density of the matching target area. Dynamically adjust the transparency, edge curvature, and internal information density of the ad container based on the semantic characteristics of the native content in the target area and the user's real-time interactive behavior; Based on the user's interactive behavior on the surface of the advertising container, the user's cognitive load is evaluated in real time. When cognitive overload is detected, the information hierarchy is folded, dynamically hiding the secondary information or all information of the advertisement.

[0007] As a further solution of the present invention: it also includes recording the entire life cycle data of the advertising container from generation to disappearance, including migration trajectory, morphological change data and user interaction behavior, the migration trajectory is the movement path of the advertising container on the screen, the morphological change data includes changes in transparency, edge curvature and information density, and the interaction behavior includes click data and sliding data; extracting the gravitational field feature patterns of the user's high interaction frequency areas in the entire life cycle data, establishing an association mapping table between semantic gravitational field parameters and user interaction habits, and iteratively optimizing the migration response speed and morphological change threshold of the advertising container through a reinforcement learning algorithm.

[0008] As a further solution of the present invention: the touch sliding behavior data includes the fingertip sliding acceleration change rate and the positional relationship, distance and angle of multiple finger contact points on the screen; the interface dwell characteristics are the user's dwell time, number of dwell times and device parameters during the dwell time on any page; the interface content semantic attributes include the emotional polarity intensity of text keywords, the visual impact index of the main color tone of the image and the motion direction consistency score of the dynamic elements of the video.

[0009] As a further solution of the present invention: the process of constructing the semantic gravitational field includes: Based on the acceleration attenuation characteristics of the user's historical sliding trajectory, the length of time the user stays on different interfaces, and the changing trend of the user's touch pressure on the screen, the user's current browsing intention trend area and trend duration are extracted; Map the text semantic attributes of the current interface into emotional gravity values, convert image visual features into color attractiveness values, and quantify video dynamic attributes into motion trend values; The extracted user interaction behavior information and interface semantic feature information are weighted and calculated to generate the gravity value of each grid area with interactive behavior and semantic features. Based on each grid area with a known gravity value, the gravity values ​​of all grid areas of the entire screen interface are calculated based on spatial interpolation technology to generate a continuous three-dimensional semantic gravity field.

[0010] As a further solution of the present invention: the initial generation process of the advertisement container is as follows: Obtain the geometric features of the blank area of ​​the current screen interface, and identify the color matching and line style of the interface elements near the blank area. Generate a corresponding outline as the container base form based on the geometric features, color matching and line style, and deform the advertising material through vector deformation technology to make it adapt to the container base form.

[0011] As a further solution of the present invention: path optimization by fluid mechanics simulation includes: When the ad container migration path passes through an area where the density of graphic elements exceeds a preset threshold, virtual viscous resistance is added to the ad container, reducing its movement speed. The curvature of the ad container migration path is dynamically adjusted to ensure that the migration trajectory always stays in an area where the density of graphic elements is below the preset threshold.

[0012] As a further solution of the present invention, the dynamic adjustment of the transparency, edge curvature and internal information density of the advertising container specifically includes: When the advertising container passes through the target area, the contrast between the target area and the background color is calculated. When the contrast is greater than a preset threshold, the edge transparency of the advertising container is increased. When the contrast is less than the preset threshold, the edge transparency of the advertising container is reduced. Each level of contrast corresponds to a level of edge transparency. The information entropy of the graphic elements in the target area is obtained, and the arrangement density of the graphic elements in the advertising container is scaled according to a set ratio based on the information entropy value. The sliding direction is obtained by monitoring the finger sliding trajectory of the user on the screen, mapped to a sliding direction vector, and the sliding direction vector is calculated to obtain the sliding direction angle θ. The graphic elements in the target area are obtained, and the visual center of gravity of the target area is calculated based on the color and brightness characteristics of the image content and the layout, size, color and other factors of the text content. The visual center of gravity before and after the user's finger sliding is marked as G1 and G2, and the visual center of gravity offset ΔG is calculated. The curvature radius R of the edge of the advertising container is calculated based on the obtained user sliding direction θ and the visual center of gravity offset ΔG, and the shape of the edge of the advertising container is adjusted based on the curvature radius R.

[0013] As a further solution of the present invention, real-time evaluation of the user's cognitive load based on the user's interactive behavior on the surface of the advertising container includes: Monitor user interaction indicators on the ad container interface, including the frequency of touch pressure changes, the time interval between pinch-to-zoom operations, and the number of times the sliding direction changes. When the interaction indicators simultaneously exceed the first dynamic threshold, it is determined to be primary cognitive overload, triggering the folding animation effect, and the marked secondary information shrinks along a circular path to the edge of the container. When the interaction indicators simultaneously exceed the second dynamic threshold, it is determined to be severe cognitive overload, and all ad information is hidden.

[0014] The present invention also includes an advertising information flow management platform for multi-dimensional visual design, which is used to implement the above-mentioned advertising information flow management method for multi-dimensional visual design, including: A data acquisition module is used to capture user interaction behaviors in real time, including touch and slide behavior data and interface dwell characteristics, and to analyze the semantic features of the current interface content, including text semantic attributes, image visual features, and video dynamic attributes; The gravitational field construction module is used to divide the screen interface into dynamic grid areas. Based on the comprehensive calculation of user interaction behavior and content semantic features, the dynamic grid areas are mapped into a three-dimensional semantic gravitational field that represents the distribution of user attention. The areas with higher user attention have higher gravitational values: The ad container generation and migration module is used to generate the initial ad container in the low-gravity area at the edge of the screen. It then drives the container to migrate to the high-gravity area based on the intensity gradient distribution of the semantic gravity field. During the migration process, it optimizes the path through fluid dynamics simulation and dynamically adjusts the container shape based on the visual density of the matching target area. An adaptive morphology adjustment module, which dynamically adjusts the transparency, edge curvature, and internal information density of the ad container based on the semantic characteristics of the native content in the target area and real-time user interaction. The cognitive load feedback module is used to evaluate the user's cognitive load in real time based on the user's interactive behavior on the advertising container surface. When cognitive overload is detected, the information hierarchy is triggered to fold and dynamically hide the secondary information or all information of the advertisement.

[0015] Beneficial effects of the present invention: This invention captures user interactions in real time and analyzes the semantic features of interface content to construct a three-dimensional semantic gravitational field to precisely locate user attention. Compared to traditional advertising placement strategies, it can more accurately deliver ads to areas of potential user attention. Regarding ad container generation and migration, it starts from the low-gravity area at the edge of the screen and optimizes the migration path and morphology based on the semantic gravitational field, significantly improving the naturalness and adaptability of ad display. It dynamically adjusts the transparency, edge curvature, and internal information density of the ad container to achieve dual adaptation to the native content in the target area and the user's real-time interactive behavior, significantly enhancing the integration of ads with the interface and reducing interference with user browsing. By monitoring user interactions on the ad container surface, it assesses cognitive load in real time and triggers information hierarchy folding in a timely manner, effectively improving the user experience. The system also records data from the entire lifecycle of the ad container, establishes a mapping table linking semantic gravitational field parameters with user interaction habits, and utilizes a reinforcement learning algorithm to iteratively optimize the ad container's migration response speed and morphology change threshold, continuously improving the intelligence and accuracy of the ad management system. This system demonstrates outstanding innovation and significant benefits in precisely targeting user attention, improving ad integration with the interface and user experience, and enabling system self-optimization. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The present invention will be further described below with reference to the accompanying drawings.

[0017] Figure 1 This is a flow chart of an advertising information flow management method applied to multi-dimensional visual design according to the present invention; Figure 2 This is a module diagram of an advertising information flow management platform applied to multiple visual designs according to the present invention. DETAILED DESCRIPTION

[0018] 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.

[0019] See also Figure 1-Figure 2 As shown, the present invention is an advertising information flow management method applied to multi-dimensional visual design, comprising the following steps: Using sensing technology and algorithms, users' interactive behaviors are captured in real time. Touch sliding behavior data records the rate of change of acceleration of the user's fingertip sliding, as well as the positional relationship, distance, and angle of multiple finger contact points on the screen. The interface dwell feature analyzes the duration and number of dwell times a user spends on any page, as well as parameters such as the device's screen brightness and network connection status during the dwell time. At the same time, semantic analysis tools are used to parse the semantic features of the current interface content. Text semantic attributes analyze the emotional polarity strength of text keywords, image visual features extract the visual impact index of the image's main color tone, and video dynamic attributes quantify the consistency score of the motion direction of dynamic elements in the video.

[0020] The screen interface is then divided into dynamic grid areas, which adjust in real time based on changes in interface content and user actions. Based on the calculation of user interaction behavior and the semantic characteristics of the content, mathematical models and algorithms are used to map the dynamic grid areas into a three-dimensional semantic gravity field that represents the distribution of user attention. Within this gravity field, areas with higher user attention have higher gravity values, providing a basis for subsequent targeted advertising.

[0021] When an ad is placed, an initial ad container is generated in a low-gravity area at the edge of the screen to avoid initially interfering with the user's viewing of the main content. Next, the ad container is guided to migrate toward high-gravity areas based on the intensity gradient of the semantic gravitational field. During this migration, fluid dynamics simulation technology is used to optimize the path, simulating fluid flow characteristics for a more natural migration path. Furthermore, the container's shape is adjusted in real time based on the visual density of the target area. If the target area has a high visual density due to the presence of numerous graphic elements, the ad container automatically adopts a simpler, more compact form.

[0022] Furthermore, the ad container is adjusted across multiple dimensions based on the semantic characteristics of the target area's native content and real-time user interactions. The contrast between the target area and the background color is calculated, and the transparency of the ad container is adjusted accordingly. Furthermore, based on the presence of graphic elements in the target area, the edge curvature and internal information density of the ad container are adjusted according to specific rules, ensuring that the ad display is more tailored to the target area and user behavior.

[0023] Finally, cognitive load is assessed based on user interactions with the ad container. This involves monitoring user interaction metrics on the ad container interface, such as the frequency of touch pressure changes, the duration between pinch-to-zoom operations, and the number of swipe direction changes. When these metrics simultaneously exceed a set threshold, cognitive overload is detected, triggering information hierarchy collapse and dynamically hiding secondary or all information within the ad, improving the user experience while viewing the ad.

[0024] A preferred embodiment of the present invention focuses on incorporating a refined recording and in-depth analysis mechanism for the entire lifecycle of ad containers. The system comprehensively and meticulously records the entire lifecycle of ad containers, from their creation to their disappearance. The migration trajectory precisely records every path the ad container takes on the screen, whether it moves in a straight line, glides in a curve, or jumps between different areas. Regarding morphological change data, close attention is paid to every adjustment in transparency, whether it shifts from high to low transparency or vice versa in different scenarios; details on changes in edge curvature, whether it becomes rounded to match a soft interface style or sharper to suit a harsh visual environment; and changes in information density, which are encrypted or simplified as the complexity of the target area's content increases. User interaction behavior is also accurately captured, including click data, including click location, frequency, and duration, to analyze user attention to different parts of the ad. Slide data records the start and end points, speed, and direction of the slide, providing insight into user intent.

[0025] Based on this rich data, we further explore the gravitational field patterns of high-user interaction frequency areas across the entire lifecycle data. Using sophisticated data mining algorithms, we explore the unique manifestations of these areas within the semantic gravitational field, such as specific gravitational value combinations and gravitational field distribution patterns. Furthermore, we construct a mapping table linking semantic gravitational field parameters to user interaction habits, closely linking semantic gravitational field parameters such as text semantic attributes, image visual features, and video dynamic attributes to user interaction habits such as clicks and swipes. Finally, using a reinforcement learning algorithm, we iteratively optimize the ad container's migration response speed and shape change threshold. Based on the feedback generated by the ad container's interaction with the user during each iteration, we dynamically adjust the migration response speed to ensure that it appears and moves appropriately in different scenarios. We also precisely optimize the shape change threshold to ensure that the ad container is displayed at the most appropriate time and in the most appropriate form.

[0026] In another preferred embodiment of the present invention, a more detailed definition of the interactive behavior data and the semantic attributes of the interface content is made. The touch sliding behavior data includes in detail the rate of change of fingertip sliding acceleration. This data can intuitively reflect the changes in force and speed during user operation, whether it is an eager and fast sliding or a slow and steady movement. At the same time, the positional relationship, distance and angle of the contact points of multiple fingers on the screen are also included. By analyzing this information, we can deeply understand the intention of the user's complex operations, such as the behavioral logic behind operations such as pinch-to-zoom and multi-finger page switching.

[0027] Interface dwell characteristics are equally diverse. The length of time a user dwells on a page directly reflects their level of attention to the content, determining whether they are simply browsing or engaging with it for extended periods. The number of dwell times reflects the importance or appeal of the page to the user; frequent dwell times often indicate that the page contains key information that users repeatedly focus on. Device parameters during dwell times, such as screen brightness, can reflect users' visual needs for page content under varying ambient lighting conditions. Network connection status can help determine whether network issues may have led to changes in user operations during browsing.

[0028] The scope of semantic attributes of interface content is also very broad. The emotional polarity strength of text keywords is used to accurately determine whether the emotion conveyed by the text is positive, negative, or neutral, whether it is enthusiastic praise, calm and objective statement, or slightly negative evaluation. The visual impact index of the image's main color tone measures the degree to which the image's main color tone attracts the user's vision. Bright and intense colors have a very different visual impact than soft and elegant colors. The motion direction consistency score of dynamic elements in the video is used to evaluate the coordination of the dynamic elements in the video content. Whether it moves in an orderly and uniform direction or moves randomly and disorganizedly, these have a significant impact on the user's viewing experience.

[0029] In another preferred embodiment of the present invention, the process of constructing the semantic gravitational field is extremely precise and scientific. First, the acceleration attenuation characteristics of the user's historical sliding trajectory are deeply explored. By carefully analyzing the law of the gradual decrease in acceleration over time in the user's past sliding operations, it is possible to accurately understand the trend of changes in the force during the user's operation, and then infer the urgency or concentration of the user's operation. At the same time, the length of time the user stays on different interfaces is comprehensively considered. The longer the stay time, the higher the user's attention to the content of the interface and the stronger the interest. In addition, pay close attention to the changing trend of the user's touch pressure on the screen. The fluctuation of pressure can reflect the user's emotional fluctuations during the browsing process and the shift of focus on interface elements. Based on this series of rich data, advanced data mining algorithms are used to extract the tendency area of ​​the user's current browsing intention, clarify which parts of the screen the user is more inclined to pay attention to at this moment; and the tendency duration, that is, to estimate the approximate length of time the user may maintain attention to these tendency areas.

[0030] Next, a quantitative conversion is performed on various content features of the current interface. The semantic attributes of the text, such as the emotional tendencies contained in the keywords in the text, are mapped to emotional gravity values ​​through a specific semantic analysis algorithm. The emotional gravity values ​​corresponding to positive text keywords are relatively high, while the emotional gravity values ​​of negative or bland text keywords are low. For image visual features, the unique attributes of the main color of the image, such as color vividness and contrast, are converted into color attraction values. The color attraction values ​​corresponding to the main colors of images with bright colors and high contrast are greater and can attract more users' attention. In terms of video dynamic attributes, the motion trajectory, speed, and direction of the dynamic elements in the video are comprehensively evaluated and quantified into motion trend values. The motion trend values ​​of dynamic elements in the video with consistent movement directions and brisk rhythms are correspondingly higher.

[0031] The extracted user interaction behavior information and interface semantic feature information are then weighted. Based on pre-set weight coefficients, a comprehensive calculation is performed on interaction behavior information, such as the preferred area and duration of the user's browsing intent, as well as interface semantic feature information such as emotional gravity value, color appeal value, and motion trend value. This weighted calculation method generates a corresponding gravity value for each grid area with interactive behavior and semantic features. Next, based on these grid areas with known gravity values, a mathematical method is used to infer the gravity values ​​of all grid areas across the entire screen interface using spatial interpolation technology. Ultimately, a continuous three-dimensional semantic gravity field is constructed that accurately reflects the distribution of user attention in three-dimensional space.

[0032] In a preferred embodiment of this embodiment, the initial generation process of the advertising container is extremely comprehensive. First, the system accurately captures the geometric features of the blank area on the current screen interface, including its shape (regular rectangle, circle, or irregular polygon); its dimensions (specific values ​​such as length, width, and radius); and its proportional relationships (e.g., the ratios between different sides or the length-to-width ratio). Simultaneously, the system discerns the color scheme of interface elements near the blank area, determining whether warm tones create a welcoming atmosphere or cool tones create a calming feeling; and the line style (smooth, flowing straight lines, sinuous curves, or angular polylines). Based on the acquired geometric features, color scheme, and line style, a complex algorithm generates a matching outline as the container base. If the blank area is nearly circular and the nearby interface elements have soft lines and warm colors, an elliptical container base may be generated. Finally, vector deformation technology is used to skillfully transform the advertising material. Vector graphics are infinitely scalable without distortion. By adjusting parameters such as the control points and line curvature of the graphics in the creative, the creative can be perfectly adapted to the container base. Regardless of the size, shape, or layout of the graphics, they can fit seamlessly with the container base to present the best display effect.

[0033] In another preferred embodiment of this embodiment, the path optimization process, performed through fluid dynamics simulation, fully considers various factors in real-world scenarios. When the ad container's migration path is expected to pass through an area where the density of graphic elements exceeds a preset threshold, the system automatically initiates an optimization mechanism. Virtual viscous resistance is added to the ad container, similar to the resistance that fluids exert on the movement of objects in real life. By simulating this viscous resistance, the ad container's movement speed is effectively reduced, preventing excessive speed from disrupting the user's normal browsing of graphic content when passing through complex areas. Simultaneously, a dynamic adjustment algorithm is used to precisely adjust the curvature of the ad container's migration path. Based on the distribution of graphic element density and the spatial layout of the surrounding area, the optimal path curvature is calculated, ensuring that the migration trajectory remains within areas where the graphic element density is below the preset threshold. This ensures that the ad container can smoothly migrate to the target area without causing visual interference in the densely packed areas where users are most focused, ensuring a smooth user experience while browsing the interface, unaffected by the ad migration process.

[0034] It is worth noting that the process of dynamically adjusting the transparency, edge curvature, and internal information density of the ad container is a highly intelligent and sophisticated operation, aimed at comprehensively improving the integration of ad display and user experience.

[0035] When the ad container passes through the target area during its migration, its first priority is to accurately calculate the contrast ratio between the target area and the background. Using advanced color analysis algorithms, the system performs a detailed comparison and calculation of various color parameters, such as hue, saturation, and brightness, of the target area's primary color and the background's color, to determine an accurate contrast ratio. The preset threshold is a reference standard based on extensive user behavior research and visual design theory. When the calculated contrast ratio exceeds the preset threshold, indicating a significant color difference between the target area and the background, the system automatically increases the transparency of the ad container's edges to help the ad container blend more naturally into the visual environment. This creates a softer, more gradual transition between the edges and the surroundings, reducing visual intrusion. Conversely, when the contrast ratio falls below the preset threshold, meaning the target area and background colors are relatively close, the system reduces the transparency of the ad container's edges to ensure legibility and enhance its visual presence. During the entire adjustment process, the contrast of each different level is carefully matched to a specific level of edge transparency. This meticulous correspondence ensures that the advertising container can achieve the best visual integration effect regardless of the color environment.

[0036] At the same time, the system further obtains the information entropy of the graphic elements in the target area. Information entropy is a key metric for measuring information uncertainty and complexity. For graphic elements in the target area, the system calculates the information entropy value by comprehensively considering factors such as the word count, vocabulary richness, and sentence complexity of the text, as well as the detail richness and number of elements in the image. Based on this information entropy value, the system scales the layout density of graphic elements within the ad container according to a pre-set ratio. If the information entropy value of the graphic elements in the target area is high, indicating that the area is information-rich and complex, the system will reduce the layout density of graphic elements within the ad container to avoid conflicts between the ad content and the native content in the target area or excessive interference with users. This will make the ad presentation more concise and clear. Conversely, if the information entropy value of the graphic elements in the target area is low, the layout density of graphic elements within the ad container can be appropriately increased to fully display the ad information without excessively increasing the cognitive burden on users.

[0037] To adjust the curvature of the ad container's edge, the system utilizes sophisticated user action monitoring technology to track the user's finger's movement trajectory on the screen in real time. This trajectory is fully recorded, starting from the initial touch point, through each intermediate point in the movement, and finally to the final exit point. This trajectory is then mapped into a sliding direction vector, and mathematical operations are performed to calculate the angle of this vector, resulting in a precise sliding direction angle θ. Simultaneously, the system conducts a comprehensive analysis of the image and text elements in the target area. For image content, it analyzes features such as color distribution, the proportion of dominant colors, and the range of brightness. For text content, it considers factors such as the layout (e.g., neat paragraph arrangement or staggered layout), as well as text size, color, and font style. By integrating these relevant image and text factors, a specialized visual center of gravity calculation model is used to determine the visual center of gravity of the target area. The visual centers before and after the user's finger slide are labeled G1 and G2, respectively, and the visual center of gravity offset ΔG is calculated using coordinate operations and other methods. Finally, the obtained user scrolling angle θ and visual center of gravity offset ΔG are substituted into a specific mathematical formula to calculate the curvature radius R of the ad container edge. For example, R = k1 × θ + k2 × | ΔG | + k3, where k1, k2, and k3 are preset coefficients. Once the curvature radius R is determined, the system uses graphics rendering technology to precisely adjust the shape of the ad container edge, ensuring that the edge shape closely matches the user's browsing flow and the visual characteristics of the target area, providing users with a smoother and more natural visual experience.

[0038] In another preferred embodiment of the present invention, real-time and accurate assessment of user cognitive load is performed based on user interaction with the ad container surface. This process plays a crucial role in optimizing the user experience. Using built-in high-precision sensors and intelligent monitoring algorithms, the system continuously and meticulously monitors a range of user interaction metrics within the ad container interface.

[0039] The frequency of touch pressure changes is a key metric. The device accurately senses every subtle change in pressure when a user's finger touches the surface of an ad container and counts the number of pressure changes per unit time. For example, if a user's touch pressure increases from an initial 5 Newtons to 8 Newtons and then decreases to 6 Newtons three times within one second, the frequency of touch pressure changes is 3 times / second. The time interval between pinch-to-zoom operations is also closely monitored. When a user pinches, the system records the timestamp of each operation's start. By calculating the difference between the timestamps of two consecutive pinch-to-zoom operations, the time interval between pinch-to-zoom operations can be calculated. For example, if the user performs the first pinch-to-zoom operation at the 5th second and the second at the 8th second, the time interval between pinch-to-zoom operations is 3 seconds. Changes in swipe direction are also accurately counted. When a user swipes on the ad container, the system determines the change in swipe direction in real time based on the coordinate changes of the user's finger's swipe trajectory. For example, if a user swipes from left to right and then suddenly changes direction from top to bottom, this counts as a swipe direction change.

[0040] When these interaction metrics simultaneously exceed the first dynamic threshold, the system determines that the user is experiencing primary cognitive overload. This first dynamic threshold isn't a fixed value; it's dynamically generated using a complex data model based on a wide range of factors, including historical user interaction data, the complexity of different ad types, and user device usage habits. For example, data analysis revealed that when presented with a typical graphic ad, most users experience some degree of cognitive difficulty when the frequency of touch pressure changes exceeds 5 times per second, the interval between pinch-to-zoom actions is less than 2 seconds, and the number of swipe direction changes exceeds 4 within 10 seconds. The system then sets these values ​​as the first dynamic threshold for this scenario. Once primary cognitive overload is determined, the system quickly triggers a collapse animation. Pre-labeled secondary ad information, such as supplementary text and less important decorative icons, slowly shrinks to the edge of the container along a carefully designed circular path. This design reduces the complexity of the ad display, helping to reduce cognitive burden on users, while preserving the ad's core message and maintaining its essential functionality.

[0041] When the interaction indicators simultaneously exceed the more stringent second dynamic threshold, the system determines that the user is in a state of severe cognitive overload. The setting of the second dynamic threshold is also based on massive data and complex algorithms. Compared with the first dynamic threshold, it is more stringent in judging user cognitive overload. For example, for complex video ads, when the frequency of touch pressure changes exceeds 8 times / second, the time interval between two-finger zoom operations is less than 1 second, and the number of changes in sliding direction exceeds 6 times within 5 seconds, the system will determine that the user is overwhelmed and will decisively hide all the information of the ad. This measure aims to minimize interference to users and prevent users from becoming disgusted with ads or even the entire application due to excessive cognitive pressure, thereby significantly improving the user's comfort and satisfaction during the browsing process.

[0042] The present invention also includes an advertising information flow management platform for multi-dimensional visual design, which is used to implement the above-mentioned advertising information flow management method for multi-dimensional visual design, including: A data acquisition module is used to capture user interaction behaviors in real time, including touch and slide behavior data and interface dwell characteristics, and to analyze the semantic features of the current interface content, including text semantic attributes, image visual features, and video dynamic attributes; The gravitational field construction module is used to divide the screen interface into dynamic grid areas. Based on the comprehensive calculation of user interaction behavior and content semantic features, the dynamic grid areas are mapped into a three-dimensional semantic gravitational field that represents the distribution of user attention. The areas with higher user attention have higher gravitational values: The ad container generation and migration module is used to generate the initial ad container in the low-gravity area at the edge of the screen. It then drives the container to migrate to the high-gravity area based on the intensity gradient distribution of the semantic gravity field. During the migration process, it optimizes the path through fluid dynamics simulation and dynamically adjusts the container shape based on the visual density of the matching target area. An adaptive morphology adjustment module, which dynamically adjusts the transparency, edge curvature, and internal information density of the ad container based on the semantic characteristics of the native content in the target area and real-time user interaction. The cognitive load feedback module is used to evaluate the user's cognitive load in real time based on the user's interactive behavior on the advertising container surface. When cognitive overload is detected, the information hierarchy is triggered to fold and dynamically hide the secondary information or all information of the advertisement.

[0043] The above is a detailed description of an embodiment of the present invention. However, the content described is only a preferred embodiment of the present invention and should not be considered to limit the scope of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of the patent coverage of the present invention.

Claims

1. An advertising information flow management method applied to multi-dimensional visual design, characterized in that: The following steps are involved: Real-time capture of user interaction behaviors, including touch and slide behavior data and interface dwell characteristics, while analyzing the semantic features of the current interface content, including text semantic attributes, image visual features, and video dynamic attributes; The screen interface is divided into dynamic grid areas. Based on the comprehensive calculation of user interaction behavior and content semantic features, the dynamic grid areas are mapped into a three-dimensional semantic gravity field that represents the distribution of user attention. The areas with higher user attention have higher gravity values. An initial ad container is generated in the low-gravity area at the edge of the screen. Based on the intensity gradient distribution of the semantic gravitational field, the container is driven to migrate to the high-gravity area. During the migration process, the path is optimized through fluid dynamics simulation, and the container shape is dynamically adjusted based on the visual density of the matching target area. Dynamically adjust the transparency, edge curvature, and internal information density of the ad container based on the semantic characteristics of the native content in the target area and the user's real-time interactive behavior; Based on the user's interactive behavior on the surface of the advertising container, the user's cognitive load is evaluated in real time. When cognitive overload is detected, the information hierarchy is folded, dynamically hiding the secondary information or all information of the advertisement.

2. The advertising information flow management method for multi-dimensional visual design according to claim 1, characterized in that: It also includes recording the entire life cycle data of the advertising container from generation to disappearance, including migration trajectory, morphological change data and user interaction behavior, wherein the migration trajectory is the movement path of the advertising container on the screen, the morphological change data includes changes in transparency, edge curvature and information density, and the interaction behavior includes click data and sliding data; extracting the gravitational field feature patterns of the user's high interaction frequency areas in the entire life cycle data, establishing an association mapping table between semantic gravitational field parameters and user interaction habits, and iteratively optimizing the migration response speed and morphological change threshold of the advertising container through a reinforcement learning algorithm.

3. The advertising information flow management method applied to multi-dimensional visual design according to claim 2, characterized in that: The touch sliding behavior data includes the rate of change of fingertip sliding acceleration and the positional relationship, distance and angle of multiple finger contact points on the screen; the interface dwell characteristics are the user's dwell time on any page, the number of dwell times and the device parameters during the dwell time; the interface content semantic attributes include the emotional polarity intensity of text keywords, the visual impact index of the main color tone of the image and the motion direction consistency score of the dynamic elements of the video.

4. The advertising information flow management method for multi-dimensional visual design according to claim 3 is characterized in that: The process of constructing the semantic gravitational field includes: Based on the acceleration attenuation characteristics of the user's historical sliding trajectory, the length of time the user stays on different interfaces, and the changing trend of the user's touch pressure on the screen, the user's current browsing intention trend area and trend duration are extracted; Map the text semantic attributes of the current interface into emotional gravity values, convert image visual features into color attractiveness values, and quantify video dynamic attributes into motion trend values; The extracted user interaction behavior information and interface semantic feature information are weighted and calculated to generate the gravity value of each grid area with interactive behavior and semantic features. Based on each grid area with a known gravity value, the gravity values ​​of all grid areas of the entire screen interface are calculated based on spatial interpolation technology to generate a continuous three-dimensional semantic gravity field.

5. The advertising information flow management method applied to multi-dimensional visual design according to claim 4 is characterized in that: The initial generation process of the advertisement container is as follows: Obtain the geometric features of the blank area of ​​the current screen interface, and identify the color matching and line style of the interface elements near the blank area. Generate a corresponding outline as the container base form based on the geometric features, color matching and line style, and deform the advertising material through vector deformation technology to make it adapt to the container base form.

6. The advertising information flow management method for multi-dimensional visual design according to claim 5, characterized in that: Path optimization through fluid dynamics simulation includes: When the ad container migration path passes through an area where the density of graphic elements exceeds a preset threshold, virtual viscous resistance is added to the ad container, reducing its movement speed. The curvature of the ad container migration path is dynamically adjusted to ensure that the migration trajectory always stays in an area where the density of graphic elements is below the preset threshold.

7. The advertising information flow management method for multi-dimensional visual design according to claim 6, characterized in that: The dynamic adjustment of the transparency, edge curvature and internal information density of the advertisement container specifically includes: When the advertising container passes through the target area, the contrast between the target area and the background color is calculated. When the contrast is greater than a preset threshold, the edge transparency of the advertising container is increased. When the contrast is less than the preset threshold, the edge transparency of the advertising container is reduced. Each level of contrast corresponds to a level of edge transparency. The information entropy of the graphic elements in the target area is obtained, and the arrangement density of the graphic elements in the advertising container is scaled according to a set ratio based on the information entropy value. The sliding direction is obtained by monitoring the finger sliding trajectory of the user on the screen, mapped to a sliding direction vector, and the sliding direction vector is calculated to obtain the sliding direction angle θ. The graphic elements in the target area are obtained, and the visual center of gravity of the target area is calculated based on the color and brightness characteristics of the image content and the layout, size, color and other factors of the text content. The visual center of gravity before and after the user's finger sliding is marked as G1 and G2, and the visual center of gravity offset ΔG is calculated. The curvature radius R of the edge of the advertising container is calculated based on the obtained user sliding direction θ and the visual center of gravity offset ΔG, and the shape of the edge of the advertising container is adjusted based on the curvature radius R.

8. The advertising information flow management method for multi-dimensional visual design according to claim 7, characterized in that: Based on the user's interactive behavior on the ad container surface, real-time assessment of the user's cognitive load includes: Monitor user interaction indicators on the ad container interface, including the frequency of touch pressure changes, the time interval between pinch-to-zoom operations, and the number of times the sliding direction changes. When the interaction indicators simultaneously exceed the first dynamic threshold, it is determined to be primary cognitive overload, triggering the folding animation effect, and the marked secondary information shrinks along a circular path to the edge of the container. When the interaction indicators simultaneously exceed the second dynamic threshold, it is determined to be severe cognitive overload, and all ad information is hidden.

9. An advertising information flow management platform for multi-dimensional visual design, used to implement the advertising information flow management method for multi-dimensional visual design according to any one of claims 1 to 8, characterized in that: include: A data acquisition module is used to capture user interaction behaviors in real time, including touch and slide behavior data and interface dwell characteristics, and to analyze the semantic features of the current interface content, including text semantic attributes, image visual features, and video dynamic attributes; The gravitational field construction module is used to divide the screen interface into dynamic grid areas. Based on the comprehensive calculation of user interaction behavior and content semantic features, the dynamic grid areas are mapped into a three-dimensional semantic gravitational field that represents the distribution of user attention. The areas with higher user attention have higher gravitational values: The ad container generation and migration module is used to generate the initial ad container in the low-gravity area at the edge of the screen. It then drives the container to migrate to the high-gravity area based on the intensity gradient distribution of the semantic gravity field. During the migration process, it optimizes the path through fluid dynamics simulation and dynamically adjusts the container shape based on the visual density of the matching target area. An adaptive morphology adjustment module, which dynamically adjusts the transparency, edge curvature, and internal information density of the ad container based on the semantic characteristics of the native content in the target area and real-time user interaction. The cognitive load feedback module is used to evaluate the user's cognitive load in real time based on the user's interactive behavior on the advertising container surface. When cognitive overload is detected, the information hierarchy is triggered to fold and dynamically hide the secondary information or all information of the advertisement.

Images