Method, device and storage medium for optimizing marketing elements using electroencephalographic and eye movement feedback

Abstract

This application discloses a method, device, and storage medium for optimizing marketing elements using EEG-eye-tracking feedback, relating to the field of marketing content optimization technology. The method includes: breaking down pre-marketing promotional materials into multiple elements; for the same element combination, obtaining the initial layout of each element as a baseline layout scheme, and generating at least two different permutation / combination schemes based on preset variation rules; obtaining EEG signals and eye-tracking data of target users when browsing each permutation / combination scheme, and calculating the interest activation intensity of each element accordingly to determine the element priority ranking result; and determining the element layout optimization scheme based on the element priority ranking result and the interest point heatmap of the pre-marketing promotional materials. This application achieves quantitative evaluation of the effect of a single element based on multi-dimensional physiological feedback, overcoming the limitations of traditional overall scheme evaluation, accurately locating key elements and their layout impact, achieving targeted optimization of marketing materials, and effectively improving the visual adaptation effect and placement value of marketing content.

Description

Technical Field

[0001] This application relates to the field of marketing content optimization technology, and in particular to a method, device and storage medium for optimizing marketing elements using EEG-eye-feedback. Background Technology

[0002] In the field of marketing plan design, to ensure marketing effectiveness, it is necessary to effectively evaluate the design quality of the marketing plan, thereby guiding its optimization and improvement. Currently, existing marketing plan quality assessment methods can only provide a comprehensive evaluation of the design quality, failing to pinpoint specific defects in the design, such as which elements or layouts have problems. This hinders efficient adjustments and optimizations, making it difficult to meet the actual needs of marketing message design optimization.

[0003] The above content is only used to help understand the technical solution of this application and does not represent an admission that the above content is prior art. Summary of the Invention

[0004] The main purpose of this application is to provide a method, device and storage medium for optimizing marketing elements using EEG-eye-feedback, aiming to solve the technical problem that existing marketing scheme quality assessment methods cannot locate the specific impact of key elements and layout in the design scheme on the publicity effect.

[0005] To achieve the above objectives, embodiments of this application provide a method for optimizing marketing elements in EEG-eye feedback, the method comprising: Break down marketing pre-launch materials into multiple atomic elements; For the same combination of elements, the initial layout of each atomic element in the combination of elements is obtained as a baseline layout scheme; According to the preset mutation rules, the layout attributes of at least one atomic element in the baseline layout scheme are adjusted sequentially to generate the first mutation scheme. Modify the mutation rule or the adjustment parameters of the mutation rule to generate a second mutation scheme; Based on the baseline layout scheme, the first variation scheme, and the second variation scheme, at least two different permutation and combination schemes are obtained; Acquire the electroencephalogram (EEG) signals and eye movement data of the target user when browsing each of the aforementioned permutation and combination schemes; The interest activation intensity of each atomic element is calculated based on the electroencephalogram (EEG) signals and the eye movement data. The element priority ranking result is determined based on the interest activation intensity of each atomic element; Based on the element priority ranking results and the interest point heatmap corresponding to the marketing preheating materials, an element layout optimization scheme is determined.

[0006] In one embodiment, the step of calculating the interest activation intensity of each atomic element based on the electroencephalogram (EEG) signal and the eye-tracking data includes: For each atomic element, the fixation duration percentage and eye movement heatmap corresponding to the atomic element are determined based on the eye movement data, thereby generating attention anchor point indicators; In response to the gaze duration exceeding a preset threshold, the EEG signal within the corresponding time window is captured. The asymmetric features of the prefrontal alpha wave were extracted from the intercepted EEG signals to generate an emotional arousal index. The ratio of beta waves to theta waves is extracted from the intercepted EEG signals to generate a click impulse index; The ratio of beta waves to prefrontal alpha waves is extracted from the intercepted EEG signals to generate an interaction willingness index. The interest activation intensity is obtained by weighting and fusing the attention anchor index, the emotional arousal index, the click impulse index, and the interaction willingness index.

[0007] In one embodiment, the step of acquiring EEG signals and eye movement data when the target user browses each of the permutation and combination schemes includes: Based on the target user's historical browsing behavior data or user profile data, determine the display order of each of the above permutation and combination schemes; The control display unit displays each of the aforementioned permutation and combination schemes in the order described. The EEG acquisition unit and the eye-tracking unit are controlled to simultaneously acquire the EEG signals and eye-tracking data of the target user within the display windows of each of the aforementioned permutation and combination schemes; The collected EEG signals and eye movement data are timestamped and then associated with the corresponding permutation and combination scheme.

[0008] In one embodiment, the step of determining the display order of each of the permutation and combination schemes based on the target user's historical browsing behavior data or user profile data includes: Based on the historical browsing behavior data, extract the target user's gaze preference distribution for the preset spatial area; Based on the user profile data, extract the interest weights of the target user for each of the atomic elements in each of the permutation and combination schemes; For each of the aforementioned permutation and combination schemes, the layout fit degree of the corresponding permutation and combination scheme is calculated based on the spatial region of each of the aforementioned atomic elements in the permutation and combination scheme and the gaze preference distribution. The permutation and combination schemes are sorted according to the layout adaptability to generate a display sequence.

[0009] In one embodiment, after determining the display order of each of the permutation and combination schemes based on the target user's historical browsing behavior data or user profile data, the EEG-eye-tracking feedback marketing element optimization method further includes: Obtain the active time period characteristics and attention decay characteristics of the target user; Based on the characteristics of the active time period, determine the initial display time when the target user's attention level is highest; Based on the attention decay characteristics, calculate the optimal time interval for the target user to browse each of the permutation and combination schemes; Starting from the initial display time, each of the permutations and combinations in the display sequence is displayed sequentially according to the optimal time interval.

[0010] In one embodiment, the step of breaking down the marketing preheating material into multiple atomic elements includes: Acquire image data of the marketing pre-launch materials; A preset semantic segmentation model is invoked to identify regions in the image data that belong to a preset set of element types. The preset set of element types includes the anchor's face region, product image region, promotional text region, live broadcast reminder region, and brand logo region. Determine the element type and bounding box coordinates corresponding to each identified region, and generate multiple atomic elements based on the element type and the bounding box coordinates.

[0011] This application embodiment also provides a marketing element optimization device for EEG-eye feedback, the marketing element optimization device for EEG-eye feedback includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, the computer program being configured to implement the steps of the marketing element optimization method for EEG-eye feedback as described above.

[0012] This application embodiment also provides a storage medium, which is a computer-readable storage medium, and stores a computer program thereon. When the computer program is executed by a processor, it implements the steps of the marketing element optimization method for EEG-eye-tracking feedback as described above.

[0013] One or more technical solutions proposed in this application have at least the following technical effects: This application breaks down marketing pre-heating materials into multiple atomic elements, refining the analysis object from the overall design scheme to the basic units constituting the design scheme. For the same element combination, based on the baseline layout scheme, the layout attributes of the atomic elements are adjusted sequentially according to preset variation rules to generate a first variation scheme, and a second variation scheme is generated by changing the variation rules or adjusting parameters, thus obtaining at least two different permutation and combination schemes. This establishes a basis for comparison between different elements and layout forms, solving the problem that traditional marketing scheme evaluation can only output the overall effect and cannot distinguish the actual role of individual elements. By collecting EEG signals and eye movement data of target users when browsing various permutation and combination schemes, and calculating the interest activation intensity of each atomic element, the neurophysiological feedback of users to each atomic element is transformed into quantifiable and comparable evaluation indicators. By determining the element priority ranking results based on the interest activation intensity of each atomic element, and combining it with the interest point heatmap corresponding to the marketing pre-heating materials, the element layout optimization scheme is comprehensively determined, and the specific elements to be optimized and their layout adjustment methods are directly output. This application solves the technical problem that existing marketing plan quality assessment methods cannot locate key elements and the specific impact of their layout on the promotional effect, and realizes the practical output of accurate identification and layout optimization of the effects of core elements in marketing design plans. Attached Figure Description

[0014] Figure 1 A flowchart illustrating an embodiment of the marketing element optimization method for EEG-eye-tracking feedback in this application; Figure 2 A flowchart illustrating Embodiment 2 of the marketing element optimization method for EEG-eye-tracking feedback in this application; Figure 3 This is a schematic diagram of the structure of the EEG-eye-tracking feedback marketing element optimization device involved in the EEG-eye-tracking feedback marketing element optimization method in the embodiments of this application.

[0015] The purpose, features, and advantages of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0016] It should be understood that the specific embodiments described herein are merely illustrative of this application and are not intended to limit this application.

[0017] To better understand the technical solution of this application, a detailed description will be provided below in conjunction with the accompanying drawings and specific implementation methods.

[0018] In the field of marketing plan design, to ensure marketing effectiveness, it is necessary to effectively evaluate the design quality of the marketing plan, thereby guiding its optimization and improvement. Currently, existing marketing plan quality assessment methods can only provide a comprehensive evaluation of the design quality, failing to pinpoint specific defects in the design, such as which elements or layouts have problems. This hinders efficient adjustments and optimizations, making it difficult to meet the actual needs of marketing message design optimization.

[0019] In view of the above problems, this application proposes a marketing element optimization method based on EEG and eye-tracking feedback. The method involves decomposing marketing pre-heating materials into multiple atomic elements; generating at least two different permutation and combination schemes for the same element combination; acquiring EEG signals and eye-tracking data of target users when browsing each of the permutation and combination schemes; calculating the interest activation intensity of each atomic element based on the EEG signals and eye-tracking data; and outputting an element layout optimization scheme based on the interest activation intensity.

[0020] This application provides a solution that breaks down marketing pre-heating materials into multiple atomic elements, refining the analysis object from an overall design scheme to the basic units constituting the design scheme. For the same element combination, based on a baseline layout scheme, the layout attributes of the atomic elements are adjusted sequentially according to preset variation rules to generate a first variation scheme, and a second variation scheme is generated by changing the variation rules or adjusting parameters, thereby obtaining at least two different permutation and combination schemes. This establishes a basis for comparison between different elements and layout forms, solving the problem that traditional marketing scheme evaluation can only output the overall effect and cannot distinguish the actual role of individual elements. By collecting EEG signals and eye movement data of target users when browsing various permutation and combination schemes, and calculating the interest activation intensity of each atomic element, the neurophysiological feedback of users to each atomic element is transformed into quantifiable and comparable evaluation indicators. By determining the element priority ranking result based on the interest activation intensity of each atomic element, and combining it with the interest point heatmap corresponding to the marketing pre-heating materials, the element layout optimization scheme is comprehensively determined, and the specific elements to be optimized and their layout adjustment methods are directly output. This application solves the technical problem that existing marketing plan quality assessment methods cannot locate key elements and the specific impact of their layout on the promotional effect, and realizes the practical output of accurate identification and layout optimization of the effects of core elements in marketing design plans.

[0021] It should be noted that the executing entity in this embodiment can be a computing service device with data processing, network communication, and program execution functions, such as a mobile phone, personal computer, tablet computer, or other terminal device; or an electronic device capable of performing the above functions, a marketing element optimization device based on EEG and eye-tracking feedback, or a marketing element optimization system. The following description uses a marketing element optimization system as an example to illustrate this embodiment and the subsequent embodiments.

[0022] Please refer to the marketing element optimization method of the EEG-eye-tracking feedback proposed in the first embodiment of this application. Figure 1 The method includes steps S10 to S90: Step S10: Decompose the marketing preheating materials into multiple atomic elements.

[0023] It's important to note that pre-marketing promotional materials refer to visual resources used for promotional purposes before a live stream or event, including live stream preview posters, short video covers, marketing pop-ups, appointment cards, or display advertisements. Atomic elements are the smallest visual units that constitute these promotional materials, including title text, the host's face, product / gift images, price / discount figures, live stream reminder buttons, countdown modules, and brand logos. Each atomic element corresponds to an independently identifiable content area, such as the host's face area, product image area, promotional text area, countdown area, brand logo area, or live stream reminder area.

[0024] Traditional marketing design typically analyzes and evaluates the entire set of materials, ultimately outputting an overall evaluation result, such as a comprehensive quality score, to determine the merits of the design. However, this approach fails to distinguish the actual impact of individual elements on user interest, nor can it pinpoint specific problems and optimization directions within the design. For example, it cannot determine whether the issue lies in the inappropriate placement of the brand logo or the lack of appeal in the presenter's facial expressions. Therefore, this embodiment breaks down the materials into multiple atomic elements, laying the foundation for subsequent precise analysis and quantitative evaluation. This allows for the identification of design problems in specific elements and their layout, and the output of actionable optimization suggestions.

[0025] As one possible implementation, step S10 includes steps S110 to S130: Step S110: Obtain image data of the marketing preheating material.

[0026] Step S120: Call the preset semantic segmentation model to identify the regions in the image data that belong to the preset element type set. The preset element type set includes the anchor's face region, the product image region, the promotional text region, the live broadcast reminder region, and the brand logo region.

[0027] Step S130: Determine the element type and bounding box coordinates corresponding to each identified region, and generate multiple atomic elements based on the element type and the bounding box coordinates.

[0028] In this embodiment, by introducing a preset semantic segmentation model to automatically identify regions in the image data of marketing pre-heating materials, key functional areas with marketing guidance functions, such as the anchor's face, product images, promotional text, broadcast reminders, and brand logos, can be accurately divided. This avoids the subjective bias and inefficiency caused by manual segmentation. At the same time, atomic elements are standardized and defined by element type and bounding box coordinates, so that each smallest visual unit has clear attributes and positional information. This provides a unified and quantifiable analytical basis for subsequent generation of various permutation and combination schemes, collection of user physiological data, and calculation of interest activation intensity, ensuring the accuracy and reproducibility of the entire optimization process.

[0029] Specifically, the marketing pre-launch materials are first converted into image data with a unified resolution and color format. The image data is then input into a semantic segmentation model, which performs pixel-by-pixel feature extraction and category prediction on the image, outputting a region mask that matches a preset set of element types. Subsequently, contour detection and coordinate calculation are performed on the masked regions to determine the minimum bounding box of each effective region and its corresponding element type. Finally, the image is cropped according to the bounding box coordinates, and each independent region is encapsulated into atomic elements containing type information, location information, and image feature information, thus completing the structured decomposition.

[0030] As an example, when the marketing pre-launch material is a lipstick live stream preview poster, the atomic elements may include the host's face area, lipstick product image, discount numerical text, countdown module for the start of the live stream, and brand logo.

[0031] Step S20: For the same element combination, obtain the initial layout of each atomic element in the element combination as the baseline layout scheme.

[0032] In this embodiment, the element combination is obtained by combining multiple atomic elements obtained from decomposition according to different selection ranges, functional categories or display requirements, which can form a variety of different element combinations. Each element combination contains different element types, element quantities and functional modules.

[0033] As an example, for the N atomic elements that have been decomposed, the system can select any subset of them to form an element combination. For example, it can select all N atomic elements to form an element combination, or select any number of atomic elements (such as N-2) to form an element combination. The same atomic element can participate in the formation of multiple different element combinations at the same time, and each element combination contains at least two atomic elements.

[0034] Once the element combination is determined, an initial layout is automatically generated based on preset layout specifications or historical optimized layout data. This involves placing each atomic element in the element combination within the default display area and assigning it standard size, color scheme, hierarchical relationship, and basic animation effects, forming a baseline layout scheme. The baseline layout scheme serves as the original reference system for all subsequent variation operations, providing a quantifiable offset benchmark for the adjustment range and direction of the variation scheme.

[0035] Step S30: According to the preset mutation rules, adjust the layout attributes of at least one atomic element in the baseline layout scheme in sequence to generate the first mutation scheme.

[0036] The preset variation rules include single attribute adjustment rules and combined attribute adjustment rules. Single attribute adjustment rules are used to change the layout attributes of a single atomic element. These layout attributes include, but are not limited to, position offset, size scaling ratio, color value, hierarchy relationship, or animation triggering method. Combined attribute adjustment rules are used to adjust the layout attributes of multiple atomic elements simultaneously.

[0037] In this embodiment, the baseline layout scheme is parametrically modified according to the preset mutation rules. The atomic element types and quantities of the original element combination remain unchanged throughout the process. Only the layout attributes of the atomic elements are adjusted in a controllable manner. The "sequential adjustment" perturbation method is adopted to ensure that the change path of each mutation to the overall layout is clear and traceable, thus generating the first mutation scheme.

[0038] Step S40: Change the mutation rule or the adjustment parameters of the mutation rule to generate a second mutation scheme.

[0039] This embodiment provides two mutation paths. The first is to modify the mutation rule itself, such as adjusting the position offset to size scaling, rotation angle, or changing element color schemes, making the second mutation scheme fundamentally different from the first in terms of mutation mechanism. The second is to change the adjustment parameters of the mutation rule, such as adjusting the displacement step of a certain atomic element to the right from 5 pixels to 15 pixels, or adjusting the scaling ratio from 1.2x to 1.5x, making the second mutation scheme the same as the first in mutation mechanism but different in mutation magnitude. These two mutation paths can be used individually or in combination to ultimately generate a second mutation scheme that differs from the first mutation scheme in mutation dimension and / or mutation magnitude, providing multiple scheme variants across dimensions and / or magnitudes for subsequent multiple comparison and verification.

[0040] Step S50: Based on the baseline layout scheme, the first variation scheme, and the second variation scheme, at least two different permutation and combination schemes are obtained.

[0041] It should be noted that permutation and combination schemes refer to multiple layout design schemes formed by differentiating the layout attributes of each atomic element in the scheme display interface, such as position, size, color scheme, hierarchical relationship, and animation effects, while keeping the element combination unchanged. Multiple permutation and combination schemes corresponding to the same element combination maintain the same element type, number, and content; the differences between schemes only stem from adjustments to layout attributes and do not involve the addition, deletion, or replacement of the elements themselves. For different element combinations, corresponding permutation and combination schemes are generated separately, ensuring that the layout optimization process for different element combinations does not interfere with each other, thereby adapting to diverse marketing pre-heating scenarios and display format requirements.

[0042] After generating the first and second variant schemes, at least two schemes are selected from the baseline layout scheme, the first variant scheme, and the second variant scheme as permutation and combination schemes generated based on the current element combination. This ensures that each group of schemes only has differences in layout attributes, providing a sample basis with controllable variables and clear comparisons for subsequent user physiological data collection and interest activation intensity calculation.

[0043] By pairing the baseline layout scheme, the first variant scheme, and the second variant scheme, multiple sets of comparative relationships can be formed. For example, the difference between the baseline layout scheme and the first variant scheme can reflect the impact of layout attribute changes under a certain variant rule on user interest. The difference between the baseline layout scheme and the second variant scheme can reflect the impact of layout changes of the same element combination under another variant dimension or variant magnitude. The difference between the first variant scheme and the second variant scheme can reflect the effect deviation between different variant paths. Based on the above multiple sets of comparisons, if a certain atomic element shows a stable and consistent difference in interest activation intensity in multiple different pairing combinations, then its importance conclusion has undergone multiple cross-validation and has good repeatability, thereby effectively improving the statistical stability of element priority ranking and layout optimization decisions.

[0044] It should be noted that this embodiment only uses the first and second mutation schemes as examples for illustration, and does not limit the number of mutation schemes. In actual applications, the mutation rules can be continuously iterated or the parameter range can be adjusted according to the marketing analysis needs to expand and generate multiple sets of mutation schemes. From the baseline layout scheme and all mutation schemes, the required number of schemes can be selected according to the preset number of schemes as the multiple permutation and combination schemes corresponding to the current element combination.

[0045] This embodiment generates at least two different permutation schemes based on all atomic elements within any given element combination, thereby constructing a hierarchical comparative analysis framework. At the element combination level, by comparing different element combinations horizontally, the element types and quantity configurations most suitable for the preset marketing objectives can be identified. Within the same element combination, since the permutation schemes differ only in the layout attributes of atomic elements, while the element types, quantities, and content remain consistent, the differences in user interest activation intensity can be uniquely attributed to changes in layout forms such as position, size, and color scheme, excluding interference from differences in the element content itself. Based on the above setup, the comparison results at different levels complement each other: the comparison at the element combination level can provide a basis for selecting the optimal content composition, while the comparison at the permutation scheme level can provide support for determining the optimal layout. The combination of the two ultimately forms a complete evaluation loop, providing a logically clear and accurately attributable experimental foundation for subsequent element priority ranking and layout optimization decisions.

[0046] As an example, when the marketing pre-launch material is an eye makeup live stream preview poster, the following three layouts can be generated for the three atomic elements containing the host's face, product image, and appointment reminder text: Layout A is a horizontal layout with the host on the left, the product on the right, and the appointment reminder text at the bottom; Layout B is a vertical layout with the host centered and enlarged, the product image in the lower left corner, and the appointment reminder text at the bottom; Layout C adds eye highlighting to the host's face and adds arrow animations around the product image pointing to the appointment reminder text.

[0047] Step S60: Obtain EEG signals and eye movement data when the target user browses each of the permutation and combination schemes.

[0048] It should be noted that EEG signals are electrical activity signals of the user's cerebral cortex collected by EEG acquisition equipment. They can objectively reflect the user's emotional arousal level, attention concentration, and interaction intentions, and include brainwave components of different frequency bands such as alpha waves, beta waves, and theta waves. Eye movement data are data related to the user's eye movements collected in real time by eye-tracking equipment, specifically including key observation information such as fixation point, single fixation duration, visual fixation trajectory, and visual saccade path.

[0049] In this embodiment, the target users are multiple testers pre-selected and recruited based on the expected audience profile of the marketing pre-heating materials. Each permutation and combination scheme is viewed sequentially by all target users. To ensure the validity and consistency of the collected data, each target user can participate in the testing process individually. The testing environment is kept stable with constant lighting and controllable ambient noise. Furthermore, the EEG acquisition equipment and eye-tracking equipment are pre-calibrated to achieve unified timing synchronization across multiple devices. Simultaneously, when a target user begins viewing any permutation and combination scheme, both types of acquisition devices are triggered to start sampling and recording. This ensures precise temporal matching between EEG signals and eye-tracking data, fully corresponding to user visual behavior and brain activity under different layout screens. This effectively avoids data misalignment caused by timing deviations, ensuring the correspondence and accuracy of the original observation data.

[0050] Furthermore, the display duration for each permutation and combination scheme is pre-set uniformly, and a fixed preset display duration can be used. Before the scheme is displayed, a reference gaze marker of a fixed duration (e.g., 1 second) is first presented in the center of the display interface. This reference gaze marker is used to guide the target user's gaze to the center of the display interface and uniformly reset the user's initial visual gaze position. At the same time, the appearance of this reference gaze marker serves as the starting trigger for synchronous acquisition by the EEG acquisition device and the eye-tracking device. After the reference gaze marker disappears, the display interface immediately loads and displays the current permutation and combination scheme. The EEG acquisition device continuously records continuous EEG waveform data, and the eye-tracking device acquires and outputs a gaze point coordinate sequence formed by arranging continuous gaze points in chronological order in real time. When the display of a single permutation and combination scheme ends, the system sends a fixed marker pulse to the EEG acquisition device and simultaneously writes the corresponding time marker into the eye movement data acquired by the eye-tracking device. This allows for subsequent segmentation and temporal alignment of the two types of physiological data through a unified marker signal, facilitating subsequent data splitting and correlation analysis of different layout schemes.

[0051] Step S70: Calculate the interest activation intensity of each atomic element based on the EEG signal and the eye movement data.

[0052] It should be noted that interest activation intensity is a comprehensive quantitative indicator used to quantify the degree of visual attraction, attention capture ability, and interaction intention guidance effect of different atomic elements on the target user.

[0053] In this embodiment, the system first performs standardized preprocessing on the acquired raw EEG signals and eye-tracking data, such as filtering to remove abnormal data caused by environmental interference, limb micro-movements, and invalid visual actions, retaining valid observation segments directly related to the permutation and combination schemes browsed by the target user. Based on the preprocessed EEG signals, the energy characteristics and fluctuation characteristics of brain waves at different frequency bands (such as alpha waves, beta waves, and theta waves) are extracted to quantify the neurophysiological state of the target user when browsing the display areas corresponding to each atomic element. This neurophysiological state includes emotional feedback and the degree of attention. Combined with the preprocessed eye-tracking data, the system statistically analyzes the number of times the target user focuses on the display areas corresponding to each atomic element, the cumulative fixation time, the visual persistence distribution, and the visual saccade path characteristics, quantifying the guiding effect of each atomic element on the user's gaze.

[0054] Furthermore, a fusion calculation logic for EEG signal correlation features and eye-tracking data correlation features is constructed. The neurophysiological feedback indicators extracted from the EEG signals are weighted and fused with the visual behavior observation indicators statistically obtained from the eye-tracking data. Based on the regional division of the permutation and combination scheme, individual indicators are calculated for the display area where each atomic element is located, effectively eliminating visual and physiological signal interference from adjacent atomic elements. Finally, the quantitative score corresponding to each atomic element is output, which represents the interest activation intensity of the corresponding atomic element. Thus, this embodiment achieves the quantitative conversion of multi-dimensional physiological data and visual behavior data, providing an objective and reliable quantitative reference for subsequent optimization of the overall layout structure.

[0055] Specifically, step S70 includes steps S710 to S760: Step S710: For each atomic element, determine the fixation duration percentage and eye movement heatmap corresponding to the atomic element based on the eye movement data, thereby generating attention anchor point indicators.

[0056] Step S720: In response to the gaze duration ratio exceeding a preset threshold, the EEG signal within the corresponding time window is captured.

[0057] Step S730: Extract the asymmetric features of the prefrontal alpha wave from the intercepted EEG signal to generate an emotion arousal index.

[0058] Step S740: Extract the ratio of β waves to θ waves from the intercepted EEG signal to generate the click impulse index.

[0059] Step S750: Extract the ratio of β wave to prefrontal α wave from the intercepted EEG signal to generate an interaction willingness index.

[0060] Step S760: The attention anchor index, the emotional arousal index, the click impulse index, and the interaction willingness index are weighted and fused to obtain the interest activation intensity.

[0061] It should be noted that the attention anchor point metric is a quantitative indicator calculated based on eye-tracking data, with data sources including eye-tracking heatmaps and fixation duration. The eye-tracking heatmap is a heat map generated from user eye movement data, which can intuitively show the density and traversal path of user gaze in different areas of marketing pre-launch materials; fixation duration is the cumulative time the user's gaze lingers within the display area corresponding to the atomic element. This attention anchor point metric uses the proportion of fixation time within the display area corresponding to the atomic element as its core, combined with the gaze concentration priority reflected by the eye-tracking heatmap, to intuitively reflect the ability of different atomic elements to capture user visual attention and their priority level, directly corresponding to the key elements that users first encounter and linger on for the longest time in a live broadcast scenario. The emotion arousal index is a quantitative indicator generated based on the asymmetric characteristics of prefrontal alpha waves, reflecting the user's emotional tendency and intensity when facing atomic elements through EEG signals. It can characterize the expectation, excitement, and other positive emotions or emotional states evoked by a particular atomic element, corresponding to the effect of materials on user emotions in a live broadcast scenario; the click impulse index is based on EEG signals... The ratio of beta waves to theta waves is a quantitative indicator that reflects the user's tendency to perform rapid, impulsive actions on atomic elements through neural activity characteristics. It can characterize the driving strength of the user's instant clicks and corresponds to the potential impulsive level of user clicks for interactive behaviors such as reservations and coupons in live streaming scenarios. The interaction willingness index is a quantitative indicator based on the ratio of beta waves to prefrontal alpha waves in EEG signals. It reflects the user's active participation tendency and operational readiness for atomic elements through cortical arousal characteristics. It can characterize the intensity of the user's deep interactive intention when watching marketing pre-heating materials and corresponds to the intrinsic driving force level of key interactive behaviors that require more cognitive input, such as users actively commenting, sharing, jumping to details, and consulting customer service in live streaming scenarios.

[0062] In this embodiment, for each atomic element, firstly, based on eye-tracking data collected by the eye-tracking device, the display area of ​​the atomic element in the solution display interface is located, the coordinate range of the display area is defined, and the display areas of other atomic elements are strictly distinguished to avoid data statistical bias caused by overlapping areas. Subsequently, combining eye-tracking heatmap and gaze duration data, the cumulative gaze duration when the target user's gaze falls on the display area corresponding to the atomic element during the current permutation and combination scheme is calculated, and the gaze duration ratio (i.e., the cumulative gaze duration of the atomic element / the total browsing time of the current permutation and combination scheme) is calculated. At the same time, based on the heat distribution characteristics of the eye-tracking heatmap, the visual attention weight of the area corresponding to the atomic element is extracted. The gaze duration ratio and the visual attention weight are weighted and fused to obtain the attention anchor point index, which is used to comprehensively quantify the ability of the atomic element to capture the user's attention. It reflects both the proportion of time the element is continuously focused on by the user and its heat distribution priority in the overall visual browsing path. The higher the index value, the stronger the attention capture effect of the atomic element.

[0063] In this embodiment, a preset gaze duration percentage threshold (hereinafter referred to as the preset threshold) is used to determine whether the user has effectively invested attention in a particular element. The system compares the gaze duration percentage of each element with the preset threshold in real time. When the gaze duration percentage of a certain element exceeds the preset threshold, it is determined that the user has effectively focused on that element, and the system then extracts the EEG signal within the corresponding time window from the original EEG signal. It is understood that EEG analysis is triggered only when the user has sufficiently focused on a particular element, avoiding signal interference during ineffective periods and thus improving the accuracy of interest activation intensity calculation.

[0064] Next, the captured EEG signals undergo secondary filtering, for example, using a bandpass filter algorithm to select alpha wave signals from the prefrontal cortex and remove interference from other frequency bands and environmental noise. Subsequently, the asymmetry characteristics of the prefrontal alpha waves are extracted, calculated as: Prefrontal alpha wave asymmetry = (Average amplitude of left prefrontal alpha waves - Average amplitude of right prefrontal alpha waves) / Average amplitude of bilateral prefrontal alpha waves. This calculation result is used to determine an emotional arousal index. A positive value indicates positive emotional arousal, while a negative value indicates negative emotional arousal; the larger the absolute value, the stronger the emotional arousal.

[0065] Furthermore, based on the EEG signals captured in the previous step for the corresponding time window, the amplitudes of beta waves and theta waves are extracted separately. The click impulse index is determined using the formula: beta wave / theta wave ratio = average beta wave amplitude / average theta wave amplitude. This click impulse index quantifies the intensity of the user's interaction intention after seeing the corresponding atomic element; the larger the ratio, the more focused the user's attention and the stronger the click impulse.

[0066] The interaction willingness index is calculated by extracting the amplitudes of beta waves and prefrontal alpha waves from EEG signals within the same time window, using the formula: Interaction Willingness Index = Average Amplitude of Beta Waves / Average Amplitude of Alpha Waves. Research shows that alpha waves are significantly suppressed when task load increases; while beta wave activation increases for behaviors requiring sustained cognitive processing and hand manipulation preparation. Therefore, a higher beta / alpha power ratio indicates higher cortical arousal and stronger interaction motivation; a lower beta / alpha power ratio indicates that the user is in a low-arousal, "blank" state with no intention to perform actions.

[0067] As an example, jointly interpreting the click impulse index (β / θ) and the interaction willingness index (β / α) can distinguish different user behavior tendencies: High β / θ and low β / α indicate high impulsivity and high arousal, meaning a group of people who are prone to blind clicking and impulsive operations; Low β / θ + high β / α indicates high restraint and high inertia, meaning very few active clicks.

[0068] This interaction willingness index directly reflects the level of users' intrinsic motivation to actively interact when viewing atomic elements, and can complement the click impulse index.

[0069] Finally, the attention anchor point metric, emotional arousal metric, click impulse index, and interaction willingness index were normalized, and weighting coefficients for each metric were set according to the preset marketing objectives, with the sum of all weighting coefficients being 1. For example, if the marketing objective focuses on attention capture, the weighting coefficient for the attention anchor point metric can be set to 0.4, and the other three metrics can each be set to 0.2; if the focus is on emotional arousal, the weighting coefficient for the emotional arousal metric can be set to 0.4, and the other three metrics can each be set to 0.2; if the focus is on user interaction conversion, the weighting coefficients for both the click impulse index and the interaction willingness index can be set to 0.3, and the weighting coefficients for both the attention anchor point metric and the emotional arousal metric can each be set to 0.2. Finally, the attention anchor point metric, emotional arousal metric, click impulse index, and interaction willingness index are weighted and fused to calculate the interest activation intensity of each atomic element. The higher this value, the stronger the visual appeal, emotional arousal effect, click-driving ability, and ability to mobilize the user's willingness to actively interact of that atomic element, providing a clear quantitative basis for subsequent layout optimization.

[0070] As a feasible implementation method, a missed fear trigger index can be further introduced. This index is extracted and quantified based on the neural activity characteristics in EEG signals that can reflect the user's sense of urgency and the psychology of missing out. Specifically, it can be constructed by analyzing the changes in event-related potentials or specific frequency band brainwave energy when users face atomic elements such as countdowns and limited-quantity reminders. It is used to characterize the triggering effect of different atomic elements on the user's sense of scarcity and the psychology of urgency, corresponding to the psychological impact assessment needs brought about by relevant materials in the live promotion scenario.

[0071] When incorporating the "fear of missing out" metric, it is simultaneously normalized. The adjusted weighting coefficients of each metric are then used to weight and fuse them, thus more comprehensively representing the overall impact of atomic elements on users. For example, the weighting coefficients for the attention anchor metric, emotional arousal metric, click impulse index, interaction willingness index, and the "fear of missing out" metric can be set to 0.3, 0.2, 0.2, 0.2, and 0.1 respectively, with the sum of these four weighting coefficients being 1.

[0072] Step S80: Determine the element priority ranking result based on the interest activation intensity of each atomic element.

[0073] In this embodiment, based on the interest activation intensity calculated for each atomic element, and using the quantified numerical value as the core sorting criterion, all atomic elements are sorted in descending order of activation intensity to clarify the importance level of each atomic element and form a complete element priority ranking result. Among them, atomic elements with higher interest activation intensity have stronger user attraction, emotional stimulation effect, and click-driving ability, and should be given higher display weight in subsequent layout optimization; atomic elements with lower interest activation intensity are visually weakened.

[0074] Step S90: Determine the element layout optimization scheme based on the element priority sorting results and the interest point heatmap corresponding to the marketing preheating materials.

[0075] It should be noted that the element layout optimization scheme refers to the optimal layout scheme generated based on the interest activation intensity and the interest point heatmap corresponding to the marketing preheating materials. It is used to guide the final display of marketing preheating materials in terms of element type, quantity, element position, size, color scheme, hierarchy and animation effects.

[0076] In this embodiment, firstly, based on the eye-tracking data collected by the eye-tracking device during the target user's browsing of various permutation and combination schemes, the density of gaze points, cumulative dwell time, and saccade trajectory of the user's gaze at various positions on the material interface are statistically analyzed. Specifically, the original eye-tracking point coordinates recorded during the browsing process are collected according to the time series, and the spatial coordinates of each gaze point are clustered to remove wandering noise. Based on this, using the spatial coordinate system of the marketing pre-heating material base map as a reference, the number of gaze points at each coordinate position is counted, and the cumulative dwell time is calculated in combination with the duration of each gaze point. At the same time, based on the spatial displacement and time difference between adjacent gaze points, a gaze saccade path is generated to characterize the user's attention shift pattern between different areas.

[0077] Then, using the marketing pre-launch material's base map as a medium, heatmaps are applied to assess the degree of eye focus in each area. Specifically, based on the statistically obtained number of views and cumulative dwell time at each coordinate location, corresponding heatmap weights are assigned to each coordinate location. A spatial interpolation algorithm is used to smoothly transition the heatmap weights between adjacent areas, and the intersection density of eye-tracking paths in each area is superimposed and merged to ultimately generate an interest point heatmap corresponding to the marketing pre-launch material. This interest point heatmap visually presents the user's attention intensity and focus priority for each area of ​​the marketing pre-launch material, clearly marking the core areas that attract the most user attention and the peripheral areas with lower attention.

[0078] Understandably, the generated permutation and combination schemes are all adapted to the unified marketing preheating material base map size and interface boundary. Only the layout attributes of atomic elements are adjusted between different permutation and combination schemes, and the basic canvas range of the material remains consistent. Therefore, it is possible to complete the drawing of the interest point heat map based on the same base map, and realize the unified comparative analysis of the visual attention area under different permutation and combination schemes.

[0079] As a feasible implementation method, the interest activation intensity of each atomic element is first normalized to obtain the weight coefficient of each atomic element in the overall layout. At the same time, the heat map of interest points is discretized into a grid, and the display interface is divided into several evaluation grids of equal size. The gaze density and cumulative dwell time in each evaluation grid are statistically analyzed to form a heat value distribution matrix.

[0080] Next, each atomic element in the element priority sorting result is matched sequentially to the unoccupied evaluation grid area with the highest value in the thermal value distribution matrix according to the weight coefficient from high to low, and the core anchor point coordinates of each atomic element are determined so that high-weight elements occupy the high-interest area first.

[0081] Subsequently, based on the preset weight-parameter mapping relationship, differentiated display parameters are assigned to atomic elements in different weight ranges to match the visual intensity of each element with its corresponding interest activation intensity. On this basis, according to the semantic relevance and visual coherence requirements between atomic elements, the anchor point coordinates are collaboratively fine-tuned, and the overall layout is adapted and verified under the boundary constraints of the marketing pre-heating material base map.

[0082] As an example, based on the element priority ranking results and combined with the visual attention distribution patterns presented by the heatmap of points of interest, high-priority atomic elements are preferentially placed in the high-attention core areas marked on the heatmap, matched with appropriate display sizes, color contrasts, and visual hierarchy relationships to enhance the exposure weight of high-value elements. Low-priority atomic elements are placed in the edge areas with lower attention, reasonably compressing their visual proportion, weakening redundant visual performance, and adjusting their stacking layers and dynamic display effects to reduce ineffective visual interference. Simultaneously, the specific spatial arrangement coordinates, visual stacking layers, and display size specifications of each atomic element are clearly defined, achieving a balance between visual harmony and information transmission efficiency in the display interface. Based on the quantitative results of multi-dimensional physiological data, the global layout is optimized in a targeted manner, ultimately forming a standardized and directly implementable element layout optimization scheme. This achieves precise matching between the visual presentation of materials and user perception feedback, improving the overall content dissemination and guidance effectiveness.

[0083] In summary, this application breaks down marketing pre-launch materials into multiple atomic elements, refining the analysis from an overall design scheme to the basic units constituting that scheme. This overcomes the limitations of overall scheme evaluation and provides a foundation for independent testing, quantitative scoring, and role tracing of individual elements. For the same element combination, based on a baseline layout scheme, the layout attributes of atomic elements are adjusted sequentially according to preset variation rules to generate a first variation scheme, and a second variation scheme is generated by changing the variation rules or adjusting parameters, thus obtaining at least two different permutation and combination schemes. This establishes a basis for comparison between different elements and layout forms, solving the problems of traditional marketing methods. Existing marketing plan evaluation methods can only output overall effects and cannot distinguish the actual role of individual elements. This application solves the problem that existing marketing plan quality evaluation methods cannot locate the specific impact of key elements and their layout on the promotional effect, achieving accurate identification of the core element effects in marketing design and a practical output of layout optimization. By collecting EEG signals and eye-tracking data from target users browsing various permutations and combinations, and calculating the interest activation intensity of each element, the application transforms the user's neurophysiological feedback to each element into quantifiable and comparable objective evaluation indicators. Furthermore, by determining the element priority ranking based on the interest activation intensity of each element and combining this with the interest point heatmap corresponding to the marketing pre-heating materials, the application comprehensively determines the element layout optimization scheme, directly outputting the specific elements to be optimized and their layout adjustment methods, achieving precise matching between the user's visual attention area and key display elements.

[0084] Based on the above embodiments of this application, in the second embodiment of this application, the content that is the same as or similar to that in the first embodiment can be referred to the above description, and will not be repeated hereafter. Based on this, please refer to... Figure 2 In the method for optimizing marketing elements using EEG-eye-tracking feedback, step S60 includes steps S610 to S640: Step S610: Determine the display order of each of the permutation and combination schemes based on the target user's historical browsing behavior data or user profile data.

[0085] Step S620: Control the display unit to display each of the above permutation and combination schemes in the order described.

[0086] Step S630: Control the EEG acquisition unit and the eye-tracking unit to simultaneously acquire the EEG signals and eye-tracking data of the target user within the display windows of each of the aforementioned permutation and combination schemes.

[0087] Step S640: Timestamp align the acquired EEG signals and eye movement data, and associate the aligned EEG signals and eye movement data with the corresponding permutation and combination scheme.

[0088] In this embodiment, historical browsing behavior data and / or user profile features of the target user are acquired. Combined with user browsing habits, visual preferences, and content acceptance patterns, the rotation display order of each permutation and combination scheme is determined differentially to avoid visual inertia, aesthetic fatigue, and interference from the experimental order caused by a fixed sorting. Following the determined rotation display order, the display unit is driven to present each permutation and combination scheme in turn, ensuring that each group of permutations and combinations is displayed independently and completely, with each group's display period distinct and non-interfering with others. Within the effective display cycle of each permutation and combination scheme, the EEG acquisition unit and eye-tracking unit are synchronously scheduled to start sampling, continuously capturing brain signals and eye movement data during the user's viewing process. Using the system timestamp as a benchmark, the asynchronously acquired EEG signals and eye movement data are time-corrected and precisely aligned to correct time offsets caused by multi-device acquisition. The time-matched brain signals and eye movement data are then associated with the currently displayed permutation and combination scheme, establishing a correspondence between the permutation and combination scheme and the user's physiological feedback data.

[0089] Further, step S610 includes steps S6110 to S6140: Step S6110: Based on the historical browsing behavior data, extract the target user's gaze preference distribution for the preset spatial area.

[0090] Step S6120: Based on the user profile data, extract the target user's interest weights for each of the atomic elements in each of the permutation and combination schemes.

[0091] Step S6130: For each of the permutation and combination schemes, calculate the layout fit degree of the corresponding permutation and combination scheme based on the spatial region where each of the atomic elements in the permutation and combination scheme is located and the gaze preference distribution.

[0092] Step S6140: Sort each of the permutation and combination schemes according to the layout adaptability to generate a display sequence.

[0093] It should be noted that historical browsing behavior data refers to interaction data recorded by eye-tracking devices or front-end embedded points when target users browse various marketing materials or similar visual interfaces over a past period (e.g., the last 30 days). This data includes gaze point coordinate sequences, gaze duration distribution, click hotspots, and eye gaze paths. This data can reflect users' visual browsing habits, such as a tendency to scan from left to right, from top to bottom, or a preference for focusing on the left-center area of ​​the screen or the facial area of ​​a person.

[0094] User profile data is a set of digital features built on the target user's registration information, historical consumption records, interest tags, interactive behaviors (clicks, likes, shares, reservations, etc.), and third-party data sources. Specifically, it may include age group, gender, region, occupation, income level, consumption preference category (such as beauty, digital products, food), content preference (such as humor, emotional, practical information), and historical click-through rate or dwell time coefficient for different element types (anchor face, product image, promotional text, action button).

[0095] When extracting the gaze preference distribution, first extract the two-dimensional coordinates (x, y) of all valid gaze points from historical browsing behavior data. Normalize the canvas size of the marketing pre-launch materials to a uniform standard (e.g., 1920 pixels wide, 1080 pixels high), and normalize all gaze point coordinates by the same proportion. Then, divide the normalized canvas into an M-row × N-column grid (e.g., M=20, N=36, each grid 96×54 pixels), and count the total gaze duration (or number of gaze points) within each grid cell to obtain the original gaze density matrix. Gaussian smooth this matrix to eliminate discrete noise, resulting in a smooth gaze density distribution. Finally, normalize it to the [0, 1] interval to form a gaze preference distribution map. The value of each grid cell in the map represents the user's historical attention intensity to that area; a higher value indicates that the user is more accustomed to gazing at that area.

[0096] When extracting interest weights, for each element type (e.g., the host's face, product image, promotional text, and live broadcast reminder button), a preference score for that user type of element is calculated based on user profile data. Specifically, this can be achieved by statistically analyzing the click-through rate, conversion rate, or average view duration of materials involving that element type in the user's historical interactions, and comparing this data with the average of all target users to derive a preference index. For example, if a user's click-through rate for promotional text is 1.5 times the average, then their interest weight for promotional text can be set to 1.5; if there is no historical data, a preset neutral value (e.g., 1.0) is used. Then, the interest weights for all element types are normalized so that their sum is 1.

[0097] As an example, data from a profile of a young female user shows that her click-through rate on the anchor's face (especially beauty anchors) is twice the average, her dwell time on product images is 1.2 times the average, and her click-through rate on promotional text is only 0.5 times the average. Therefore, the extracted interest weights can be set as follows: anchor's face 0.5, product images 0.3, promotional text 0.1, and live broadcast reminder area 0.1 (taking the lower values ​​since there is no specific preference).

[0098] When calculating layout fit, for the current permutation and combination scheme, the bounding box (i.e., the spatial area occupied) of each atomic element is known. Let S be the set of grid cells covered by all atomic elements under this scheme. For each grid cell g in S, its gaze preference distribution value is denoted as Pref(g), and the user interest weight corresponding to the element type e to which this grid cell belongs is denoted as Weight(e). The formula for calculating the layout fit score of this permutation and combination scheme is as follows:

[0099] Where g represents any grid cell covered by an atomic element in the current permutation scheme; S represents the set of all grid cells covered by atomic elements; |S| represents the total number of grid cells in set S; Pref(g) represents the gaze preference distribution value corresponding to grid cell g, ranging from 0 to 1; type(g) represents the element type covering grid cell g; Weight(type(g)) represents the user interest weight of the corresponding element type; type(g) represents the element type covering grid cell g (if a grid is covered by multiple elements, the element type with the highest weight is taken; or weighted according to area ratio). The higher the layout fit score, the better the layout of the permutation scheme matches the user's gaze preference and element interest preference.

[0100] In this embodiment, the user's spatial gaze preference distribution is first mined and extracted based on the target user's historical browsing behavior data, clarifying the user's preferred visual attention areas and browsing habits. Simultaneously, user profile data is combined to quantify the target user's interest weights for different types of atomic elements, forming a quantitative basis for element preference. For each permutation and combination scheme, the layout adaptability of the corresponding scheme is calculated by comparing it with the actual spatial area of ​​each atomic element in the scheme and the acquired user gaze preference distribution, objectively measuring the degree of fit between the overall screen structure and the user's visual habits. Finally, the calculated layout adaptability values ​​are used to sort the schemes in descending or ascending order, generating a differentiated scheme display sequence. This determines the order in which each permutation and combination scheme is displayed, reducing the experiential interference and visual fatigue caused by a fixed playback order, and improving the matching degree and reference value of the test feedback from multiple permutation and combination schemes.

[0101] Furthermore, step S610 is followed by steps S611 to S614: Step S611: Obtain the active time period characteristics and attention decay characteristics of the target user.

[0102] Step S612: Based on the active time period characteristics, determine the initial display time when the target user's attention level is highest.

[0103] Step S613: Based on the attention decay characteristics, calculate the optimal time interval for the target user to browse each of the permutation and combination schemes.

[0104] Step S614: Starting from the initial display time, display each of the permutation and combination schemes in the display sequence in sequence according to the optimal time interval.

[0105] In this embodiment, based on the target user's historical operation records and browsing behavior data, the corresponding active time period characteristics and attention decay characteristics are analyzed and extracted to understand the rhythmic patterns of the user's daily visual viewing and the continuous changes in attention. Based on the active time period characteristics, time intervals where the user's concentration and cognitive state are relatively better are selected, and these time intervals are set as the initial display time for the overall presentation of multiple permutation and combination schemes. Based on attention decay characteristics, the pattern of energy decline after continuous viewing of visual content is quantitatively analyzed, and the optimal time interval between adjacent permutation and combination schemes is calculated accordingly. Using the determined initial display time as the starting node, and then according to the optimal time interval, each permutation and combination scheme within the display sequence is presented sequentially and orderly to match the user's natural attention fluctuation rhythm, alleviate energy decline and sensory fatigue caused by continuous viewing, maintain the user's stable state during the viewing of a single scheme, and ensure the stability and effectiveness of subsequent physiological data collection.

[0106] Specifically, by statistically analyzing the time distribution data of target users' historical browsing behavior, we can determine the content browsing duration, page dwell time, content interaction frequency, and high-frequency browsing behavior concentration intervals within different time periods, thereby summarizing the characteristics of active time periods. Simultaneously, by combining the changes in browsing duration across multiple consecutive content segments, the frequency of interface switching, and the continuous decay trend of operation response speed, we can quantify the decline in user state and the rate of attention loss after prolonged exposure to visual content. Based on this, we can fit attention decay characteristics, achieving automated analysis and accurate extraction of the above two types of features.

[0107] Furthermore, the time interval between adjacent permutations and combinations does not need to be a uniform fixed value, but can be dynamically adjusted based on the user's real-time attention fluctuations. This embodiment monitors the changes in the user's physiological feedback during the viewing of each permutation and combination in real time, and combines this with the real-time level of attention consumption to flexibly adjust the transition time between two sets of permutations and combinations, achieving adaptive adaptation of the interval time, further maintaining a stable user observation state, and weakening the perceptual bias caused by long-term observation.

[0108] As a feasible implementation method, a time-series prediction model can be used to dynamically predict changes in the user's attention state at different viewing stages by combining the target user's historical attention decay data with real-time physiological feedback (such as eye movement fixation duration and EEG signal fluctuations). This allows for dynamic adjustment of the time intervals between various permutation and combination schemes, making the time interval settings more closely match the user's real-time attention fluctuations. This avoids attention redundancy or insufficiency caused by fixed intervals and further ensures the stability of the user's viewing state.

[0109] This embodiment precisely matches the behavioral habits of target users, optimizes the display logic and data collection specifications of each permutation and combination scheme, determines the display order of each permutation and combination scheme based on users' historical behavior data, and simultaneously initiates EEG and eye-tracking data collection. Asynchronous deviations are eliminated through timestamp calibration, achieving precise binding between physiological data and corresponding permutation and combination schemes. Furthermore, it combines user profiles and spatial preferences to calculate layout adaptability and generate display sequences. With dynamic time interval adjustment, it avoids the experience deviation caused by fixed intervals. Through time sequence alignment and data association, it achieves precise quantification of the role of atomic elements, eliminates the limitations of traditional overall scheme judgment, ensures the independence of single scheme display, and realizes the standardization and traceability of single atomic element role analysis and layout optimization. This promotes precise adjustment of marketing material layout, improves the adaptability of marketing materials to user needs, and enhances dissemination efficiency, ultimately achieving precise optimization of key elements and layout of marketing materials.

[0110] This application provides a marketing element optimization device for EEG-eye feedback. The marketing element optimization device for EEG-eye feedback includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the marketing element optimization method for EEG-eye feedback in the above embodiment 1.

[0111] The following is for reference. Figure 3 The diagram illustrates a structural schematic of a marketing element optimization device suitable for implementing EEG-eye-tracking feedback according to embodiments of this application. The marketing element optimization device for EEG-eye-tracking feedback in embodiments of this application may include various hardware and software components for implementing marketing element optimization methods for EEG-eye-tracking feedback. Figure 3 The EEG-eye-feedback marketing element optimization device shown is merely an example and should not impose any limitations on the functionality and scope of use of the embodiments of this application.

[0112] like Figure 3As shown, the EEG-eye-feedback marketing element optimization device may include a processing unit 1001 (e.g., a central processing unit, a graphics processing unit, etc.) that can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 1002 or a program loaded from a storage device 1003 into a random access memory (RAM) 1004. The RAM 1004 also stores various programs and data required for the operation of the EEG-eye-feedback marketing element optimization device. The processing unit 1001, the ROM 1002, and the RAM 1004 are interconnected via a bus 1005. An input / output (I / O) interface 1006 is also connected to the bus. Typically, the following systems can be connected to the I / O interface 1006: input devices 1007 including, for example, touchscreens, touchpads, keyboards, etc.; output devices 1008 including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; storage devices 1003 including, for example, magnetic tapes, hard disks, etc.; and communication devices 1009. Communication device 1009 allows the EEG-eye-tracking marketing element optimization device to communicate wirelessly or wiredly with other devices to exchange data. While the figure shows EEG-eye-tracking marketing element optimization devices with various systems, it should be understood that implementation or possession of all the systems shown is not required. More or fewer systems may be implemented alternatively.

[0113] Specifically, according to the embodiments disclosed in this application, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments disclosed in this application include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication device, or installed from storage device 1003, or installed from read-only memory 1002. When the computer program is executed by processing device 1001, it performs the functions defined in the methods of the embodiments disclosed in this application.

[0114] The EEG-eye-tracking feedback marketing element optimization device provided in this application, employing the EEG-eye-tracking feedback marketing element optimization method in the above embodiments, can solve the technical problem that existing marketing plan quality assessment methods cannot locate the specific impact of key elements and layout in the design scheme on the promotional effect. Compared with the prior art, the beneficial effects of the EEG-eye-tracking feedback marketing element optimization device provided in this application are the same as the beneficial effects of the EEG-eye-tracking feedback marketing element optimization method provided in the above embodiments, and other technical features in this EEG-eye-tracking feedback marketing element optimization device are the same as those disclosed in the previous embodiment method, and will not be repeated here.

[0115] It should be understood that the various parts disclosed in this application can be implemented using hardware, software, firmware, or a combination thereof. In the description of the above embodiments, specific features, structures, materials, or characteristics can be combined in any suitable manner in one or more embodiments or examples.

[0116] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0117] This application provides a computer-readable storage medium having computer-readable program instructions (i.e., a computer program) stored thereon, which are used to execute the marketing element optimization method of EEG-eye-tracking feedback in the above embodiments.

[0118] The computer-readable storage medium provided in this application may be, for example, a USB flash drive, but is not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to: electrical connections having one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this embodiment, the computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, device, or apparatus. The program code contained on the computer-readable storage medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, radio frequency (RF), etc., or any suitable combination thereof.

[0119] The aforementioned computer-readable storage medium may be included in the EEG-Eye Feedback Marketing Element Optimization Device; or it may exist independently and not be assembled into the EEG-Eye Feedback Marketing Element Optimization Device.

[0120] The aforementioned computer-readable storage medium carries one or more programs that, when executed by the EEG-eye-feedback marketing element optimization device, cause the EEG-eye-feedback marketing element optimization device to: decompose marketing preheating materials into multiple atomic elements; generate at least two different permutation and combination schemes for the same element combination; acquire EEG signals and eye movement data when the target user browses each of the permutation and combination schemes; calculate the interest activation intensity of each atomic element based on the EEG signals and the eye movement data; and output an element layout optimization scheme based on the interest activation intensity.

[0121] Computer program code for performing the operations of this application can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, and C++, as well as conventional procedural programming languages ​​such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., using an Internet connection provided by an Internet service).

[0122] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, may be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0123] The modules described in the embodiments of this application can be implemented in software or hardware. The names of the modules do not necessarily limit the functionality of the unit itself.

[0124] The readable storage medium provided in this application is a computer-readable storage medium that stores computer-readable program instructions (i.e., a computer program) for executing the above-described EEG-eye-tracking feedback marketing element optimization method. This solves the technical problem that existing marketing scheme quality assessment methods cannot pinpoint the specific impact of key elements and layout in the design scheme on the promotional effect. Compared with the prior art, the beneficial effects of the computer-readable storage medium provided in this application are the same as those of the EEG-eye-tracking feedback marketing element optimization method provided in the above embodiments, and will not be repeated here.

[0125] This application provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the marketing element optimization method for EEG-eye-tracking feedback as described above.

[0126] The computer program product provided in this application can solve the technical problem that existing marketing plan quality assessment methods cannot pinpoint the specific impact of key elements and layout in the design scheme on the promotional effect. Compared with the prior art, the beneficial effects of the computer program product provided in this application are the same as the beneficial effects of the EEG-eye-tracking feedback marketing element optimization method provided in the above embodiments, and will not be repeated here.

[0127] The above are merely preferred embodiments of this application and do not limit the patent scope of this application. Any equivalent structural or procedural transformations made using the content of this application's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent scope of this application.

[0128] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or system that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or system. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or system that includes that element.

[0129] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method.

[0130] The above are merely preferred embodiments of this application and do not limit the patent scope of this application. Any equivalent structural or procedural transformations made using the content of this application's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of this application.

Claims

1. A method for optimizing marketing elements using EEG-eye-feedback, characterized in that, The marketing element optimization methods for EEG-eye feedback include: Break down marketing pre-launch materials into multiple atomic elements; For the same combination of elements, the initial layout of each atomic element in the combination of elements is obtained as a baseline layout scheme; According to the preset mutation rules, the layout attributes of at least one atomic element in the baseline layout scheme are adjusted sequentially to generate the first mutation scheme. Modify the mutation rule or the adjustment parameters of the mutation rule to generate a second mutation scheme; Based on the baseline layout scheme, the first variation scheme, and the second variation scheme, at least two different permutation and combination schemes are obtained; Acquire the electroencephalogram (EEG) signals and eye movement data of the target user when browsing each of the aforementioned permutation and combination schemes; The interest activation intensity of each atomic element is calculated based on the electroencephalogram (EEG) signals and the eye movement data. The element priority ranking result is determined based on the interest activation intensity of each atomic element; Based on the element priority ranking results and the interest point heatmap corresponding to the marketing preheating materials, an element layout optimization scheme is determined.

2. The electroencephalographic eye feedback marketing element optimization method of claim 1, wherein, The step of calculating the interest activation intensity of each atomic element based on the electroencephalogram (EEG) signal and the eye movement data includes: For each atomic element, the fixation duration percentage and eye movement heatmap corresponding to the atomic element are determined based on the eye movement data, thereby generating attention anchor point indicators; In response to the gaze duration exceeding a preset threshold, the EEG signal within the corresponding time window is captured. The asymmetric features of the prefrontal alpha wave were extracted from the intercepted EEG signals to generate an emotional arousal index. The ratio of beta waves to theta waves is extracted from the intercepted EEG signals to generate a click impulse index; The ratio of beta waves to prefrontal alpha waves is extracted from the intercepted EEG signals to generate an interaction willingness index. The interest activation intensity is obtained by weighting and fusing the attention anchor index, the emotional arousal index, the click impulse index, and the interaction willingness index.

3. The electroencephalographic eye movement feedback marketing element optimization method of claim 1, wherein, The step of acquiring the EEG signals and eye movement data of the target user when browsing each of the permutation and combination schemes includes: Based on the target user's historical browsing behavior data or user profile data, determine the display order of each of the above permutation and combination schemes; The control display unit displays each of the aforementioned permutation and combination schemes in the order described. The EEG acquisition unit and the eye-tracking unit are controlled to simultaneously acquire the EEG signals and eye-tracking data of the target user within the display windows of each of the aforementioned permutation and combination schemes; The collected EEG signals and eye movement data are timestamped and then associated with the corresponding permutation and combination scheme.

4. The electroencephalographic eye feedback marketing element optimization method of claim 3, wherein, The step of determining the display order of each of the permutation and combination schemes based on the target user's historical browsing behavior data or user profile data includes: Based on the historical browsing behavior data, extract the target user's gaze preference distribution for the preset spatial area; Based on the user profile data, extract the interest weights of the target user for each of the atomic elements in each of the permutation and combination schemes; For each of the aforementioned permutation and combination schemes, the layout fit degree of the corresponding permutation and combination scheme is calculated based on the spatial region of each of the aforementioned atomic elements in the permutation and combination scheme and the gaze preference distribution. The permutation and combination schemes are sorted according to the layout adaptability to generate a display sequence.

5. The electroencephalographic eye movement feedback marketing element optimization method of claim 4, wherein, After the step of determining the display order of each of the permutation and combination schemes based on the target user's historical browsing behavior data or user profile data, the EEG-eye-tracking feedback marketing element optimization method further includes: Obtain the active time period characteristics and attention decay characteristics of the target user; Based on the characteristics of the active time period, determine the initial display time when the target user's attention level is highest; Based on the attention decay characteristics, calculate the optimal time interval for the target user to browse each of the permutation and combination schemes; Starting from the initial display time, each of the permutations and combinations in the display sequence is displayed sequentially according to the optimal time interval.

6. The electroencephalographic eye feedback marketing element optimization method of claim 1, wherein, The step of breaking down marketing pre-heating materials into multiple atomic elements includes: Acquire image data of the marketing pre-launch materials; A preset semantic segmentation model is invoked to identify regions in the image data that belong to a preset set of element types. The preset set of element types includes the anchor's face region, product image region, promotional text region, live broadcast reminder region, and brand logo region. Determine the element type and bounding box coordinates corresponding to each identified region, and generate multiple atomic elements based on the element type and the bounding box coordinates.

7. An electroencephalographic eye movement feedback marketing element optimization device, characterized by, The EEG-eye-tracking feedback marketing element optimization device includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, the computer program being configured to implement the steps of the EEG-eye-tracking feedback marketing element optimization method as described in any one of claims 1 to 6.

8. A storage medium, characterized by The storage medium is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by a processor, it implements the steps of the marketing element optimization method of EEG eye-tracking feedback as described in any one of claims 1 to 6.

Images