Intelligent Dynamic Advertising Content Delivery Method Based on User Emotion Recognition

Abstract

The present invention relates to an intelligent method for dynamically delivering advertising content based on user emotion recognition; the method collects facial image data of a user watching an advertisement in real time through a camera device, and inputs the data into a computer vision emotion analysis model to generate line-of-sight focus data and a user experience score, where the user experience score includes interest level, concentration level, and emotional tendency; based on this score, the advertising content with the highest matching degree with the current user state is selected from a preset advertising content library, and according to the line-of-sight focus data recognized by the emotion analysis model, the visual layout of the advertising content and the positions of key information elements are dynamically adjusted; in addition, the method also adjusts the presentation mode of the advertisement according to the real-time change of the user experience score to enhance the personalization and interactive experience of the advertisement; the present invention can improve the precise delivery of advertising content and user attention, and is applicable to application scenarios such as digital advertising screens, e-commerce platforms, and intelligent retail terminals.

Description

Technical Field

[0001] The present invention relates to the technical field of advertisement delivery, and in particular to a method for intelligent delivery of dynamic advertisement content based on user emotion recognition. Background Art

[0002] In the prior art, advertising delivery mainly relies on personalized recommendations based on user historical behavior, interest tags or demographic characteristics, such as analyzing the preferences of users through browsing history, purchase data or social media interactions to match the corresponding advertising content. In addition, some advertising delivery systems use static display or fixed playback mode, and the content and presentation method remain unchanged regardless of whether the user is interested in the advertisement. In recent years, the development of computer vision technology has enabled advertising systems to use cameras to capture users' facial expressions and eye tracking, thereby analyzing users' emotional states to optimize the push strategy of advertising content.

[0003] The existing advertising delivery system still has certain limitations in terms of accuracy and real-time adjustment capabilities. On the one hand, the recommendation method based on historical behavior cannot timely reflect the user's immediate emotions and attention during the ad viewing process, resulting in some delivery content failing to match the user's current interests. On the other hand, the existing advertising delivery technology that uses computer vision for emotion recognition is usually only used for ad selection, and fails to dynamically adjust key elements such as the visual layout of the ad and the way the information is presented to further enhance the user experience. In addition, the adjustment of ad content often lacks real-time response capabilities and is difficult to adapt to the rapid changes in user emotions, thus affecting the effectiveness of the ad.

[0004] In order to solve the above problems, the present invention provides a method for intelligent delivery of dynamic advertising content based on user emotion recognition. Summary of the invention

[0005] The present application provides a method for intelligent delivery of dynamic advertising content based on user emotion recognition to improve the accurate delivery of advertising content and user attention.

[0006] The present application provides a method for intelligent delivery of dynamic advertising content based on user emotion recognition, comprising:

[0007] The facial image data of the user when watching the advertisement is collected in real time through a camera device, including facial expressions, eye movement trajectories, gaze lingering positions and lingering time;

[0008] Inputting the facial image data into a computer vision sentiment analysis model to generate gaze focus data and user experience scores including interest, concentration, and emotional tendency;

[0009] Select the advertisement content with the highest matching degree with the current user status from a preset advertisement content library according to the user experience score;

[0010] Based on the line-of-sight focus data recognized by the computer vision emotion analysis model, dynamically adjust the visual layout and the positions of key information elements of the selected advertisement content;

[0011] Dynamically adjust the presentation mode of the advertisement content according to the real-time change of the user experience score, including content type, picture-text ratio, and information density.

[0012] The beneficial effects of the technical solution provided by this application include:

[0013] (1) Based on the real-time emotion recognition result of the user, this invention selects the advertisement content with the highest matching degree with the current user status from a preset advertisement content library, thereby improving the relevance of the advertisement, enhancing the user's viewing interest, avoiding ineffective or inefficient advertisement placement, and improving the advertisement conversion rate. (2) By analyzing the user's line-of-sight focus data and emotion tendency, dynamically adjust the visual layout and the positions of key information elements of the advertisement content, making the advertisement content more in line with the user's visual habits and the attention concentration area, thereby improving the readability and attractiveness of the advertisement, and enhancing the user's immersion and engagement. (3) This invention can dynamically adjust the presentation mode of the advertisement according to the real-time change of the user experience score, including content type, picture-text ratio, and information density, enabling the advertisement to adapt to the change of the user's emotion and attention, ensuring that the advertisement information is conveyed in the optimal way, and thus improving the advertisement dissemination effect. Brief Description of the Drawings

[0014] Figure 1 is a flowchart of a method for intelligent placement of dynamic advertisement content based on user emotion recognition provided by the first embodiment of this application. Detailed Embodiment

[0015] Many specific details are set forth in the following description to facilitate a full understanding of this application. However, this application can be implemented in many other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the connotation of this application. Therefore, this application is not limited by the specific embodiments disclosed below.

[0016] The first embodiment of this application provides a method for intelligent placement of dynamic advertisement content based on user emotion recognition. Please refer to Figure 1 , which is a schematic diagram of the first embodiment of this application. The following will be described in detail with reference to Figure 1 a method for intelligent placement of dynamic advertisement content based on user emotion recognition provided by the first embodiment of this application.

[0017] Step S101: Real-time collect the facial image data of the user when watching the advertisement, including facial expressions, eye movement trajectories, gaze fixation positions, and fixation durations.

[0018] In step S101, the facial image data of the user when watching the advertisement is collected in real time to provide a basis for emotion recognition and advertisement dynamic adjustment. The camera device can use a high-resolution camera, an infrared camera, or a camera device with depth perception ability to ensure that the user's facial expressions, eye movement trajectories, gaze fixation positions, and fixation durations can be accurately captured. The installation position of the camera device can be optimized according to the application scenario. For example, in front of an indoor digital advertising screen, the camera device can be embedded above or below the display screen to obtain the best user facial perspective; on self-service terminals or intelligent retail devices, the camera device can be installed on the top of the screen to achieve frontal collection of the user's face.

[0019] The collection of facial image data uses high-frame-rate video capture technology to reduce motion blur and improve the recognition accuracy of key facial features. The collected data includes, but is not limited to, facial expression features such as eyebrow curvature, mouth corner curve, eyelid opening and closing states, etc., which can be used to judge the user's emotional state. The eye movement trajectory is determined by detecting the continuous change of the pupil center position. Usually, computer vision technologies such as optical flow method and Hough transform are used for recognition, and at the same time, the Kalman filter or LSTM (Long Short-Term Memory Network) is combined to smooth the trajectory data to reduce the acquisition error. The gaze fixation position is determined based on the mapping relationship between the gaze direction and the screen content area. The camera device can combine the screen coordinate system and three-dimensional depth information to calculate the specific position of the user's gaze landing point through regression analysis or geometric transformation. In addition, the measurement of the fixation duration is based on the persistence of the gaze focus. If the user's gaze stays in a certain advertisement area for more than the set threshold, then this area can be recorded as the user's high-attention area.

[0020] To improve the robustness of facial data collection, step S101 can also combine multi-frame image fusion technology to improve the accuracy of facial feature detection. In low-light environments, infrared imaging or HDR (High Dynamic Range) image enhancement technology can be used to ensure the visibility of facial expressions and gaze features. For possible facial occlusion problems, such as when the user wears glasses, a hat, or a mask, a deep learning-based occlusion compensation algorithm can be used, such as using a pre-trained convolutional neural network (CNN) to predict the facial features of the occluded part.

[0021] In addition, in order to ensure the real-time nature of the data, the camera device needs to have high-speed data transmission capabilities, and can use USB3.0, PCIe or MIPI transmission protocols to reduce acquisition delays. On the processing side, FPGA (field programmable gate array) or GPU can be used for parallel accelerated computing to improve the processing efficiency of facial image data and ensure the real-time nature of user emotion recognition. For privacy protection, the collected facial data can be processed by edge computing, that is, feature extraction is completed on the local device to avoid transmitting complete image data, so as to reduce the risk of user privacy leakage.

[0022] Through the above steps, the user's facial expressions, eye movement trajectories, gaze position and duration can be obtained efficiently and accurately, providing reliable data support for subsequent sentiment analysis and dynamic adjustment of advertising content.

[0023] Furthermore, the real-time collection of facial image data of the user when watching the advertisement by means of a camera device includes:

[0024] After detecting that the user's face enters the advertisement viewing area, the exposure parameters of the camera device are dynamically adjusted based on active light adaptation imaging technology to adapt to different ambient lighting conditions;

[0025] Combined with an adaptive frame rate control algorithm, the sampling frame rate is increased when the user's eye movement frequency is high to accurately capture the instantaneous attention shift, and the sampling frequency is reduced when the user's gaze remains stable to reduce data redundancy and optimize computing efficiency.

[0026] After detecting that the user's face has entered the ad viewing area, the system first uses a face detection algorithm to determine the user's position, and dynamically adjusts the exposure parameters of the camera device through active light adaptation imaging technology to adapt to different ambient lighting conditions. The adjustment of exposure parameters is based on a real-time light intensity analysis module, which detects the current light level through the ambient light sensor of the camera device and calculates the optimal exposure value so that the facial image can maintain high contrast and clarity in both strong light and low light environments. For example, in a strong light environment outdoors, the system will reduce the exposure time and optimize the contrast enhancement algorithm to reduce the loss of details caused by overexposure; in a low light environment, the system will increase the gain parameter or enable infrared fill light to ensure that facial features are still clearly visible.

[0027] To ensure high-precision tracking of the user's eye movement trajectory and gaze fixation position, the system combines an adaptive frame rate control algorithm to dynamically adjust the frame rate of image acquisition according to the user's eye movement frequency. In the case where the user's gaze moves rapidly or the eye movement trajectory changes drastically, the system will automatically increase the sampling frame rate to capture the attention changes within a short period. For example, when the user's eyes quickly jump between multiple regions of the screen, the system will increase the data acquisition density to accurately record the process of gaze transfer. When the user's gaze stays in a fixed area for a long time and the eye movement changes little, the system will reduce the sampling frame rate to reduce data redundancy and computational burden, while ensuring that the recorded fixation time data is still accurate. This adaptive frame rate control mechanism, combined with historical eye movement pattern analysis, enables the system to collect data in the optimal way at different stages of the user watching the advertisement, ensuring both the accuracy of tracking and the utilization rate of computing resources.

[0028] Through the above method, the system can stably obtain facial expressions, eye movement trajectories, and gaze fixation information under different lighting environments and changes in user viewing behaviors, providing high-quality data support for subsequent emotion analysis and advertisement content optimization, and improving the accuracy of advertisement personalized matching and the effect of user experience optimization.

[0029] Step S102: Input the facial image data into a computer vision emotion analysis model to generate gaze focus data and user experience scores including interest, concentration, and emotional tendency.

[0030] In step S102, the facial image data is input into a computer vision emotion analysis model to analyze the user's emotional state and generate gaze focus data and user experience scores. This process first preprocesses the collected facial image data, including image denoising, illumination compensation, facial key point detection, and normalization processing, to improve the stability and accuracy of subsequent analysis. Facial key point detection can use a variety of deep learning models, such as detection based on convolutional neural networks (CNNs) or self-supervised learning models, to identify the positions of facial feature points such as eyebrows, eyes, and corners of the mouth, so as to extract features related to the user's expression.

[0031] In the facial expression analysis process, the computer vision sentiment analysis model uses the trained expression classification network to classify the user's expression state, and combines the expression change sequence to predict the emotional trend. For example, if the user's mouth corners are raised and eyes are wide open, it can be determined that the user's emotions tend to be positive, while if frowning and drooping mouth corners are detected, it may indicate dissatisfaction or low interest. In addition, for the user's eye movement trajectory and gaze retention data, the sentiment analysis model tracks the user's gaze direction and combines the screen area division information to calculate the user's gaze focus and determine the distribution of their attention on the advertising content. This process can predict the user's gaze point through corneal reflection detection (PCCR) or deep learning regression methods, and use dynamic time warping (DTW) or Kalman filtering methods to smooth the gaze trajectory to improve tracking accuracy.

[0032] The generation of user experience scores is a comprehensive value calculated based on multi-dimensional emotional features, including key indicators such as interest, concentration and emotional tendency. Interest can be calculated by the user's gaze dwell time, the frequency of facial expression changes and the blinking rate. Longer gazes accompanied by positive expressions such as smiles may indicate higher interest, while frequent blinking or rapid gaze aversion may indicate lower interest. The calculation of concentration usually involves the user's eye movement stability and changes in the area of ​​focus. For example, if a user stares at a specific advertising element for a long time and the eye movement trajectory deviates less, it indicates that the user's concentration is high, while if the gaze changes frequently, it indicates that the concentration is low. Emotional tendency combines expression recognition and semantic sentiment analysis technology, based on expression classification results and time series analysis, to determine whether the user is in a positive, negative or neutral emotional state, and uses time series models such as long short-term memory networks (LSTM) to predict the trend of emotional changes.

[0033] In the above analysis process, pre-trained deep learning models can be used, such as expression recognition networks based on ResNet and VGG, and emotional trend analysis models based on Transformer structures, to improve the accuracy and generalization of calculations. In addition, the calculation of user experience scores can adopt a weighted scoring algorithm, which assigns different weights according to the user's facial expressions, sight data, and attention characteristics to calculate a comprehensive score, thereby more accurately reflecting the user's advertising experience status. In order to improve real-time performance, computer vision sentiment analysis models can be deployed on edge computing devices so that data processing is performed locally to reduce the latency of cloud processing and enhance user privacy protection capabilities. Ultimately, the generated user experience score will serve as the basis for subsequent advertising content selection and dynamic adjustment to optimize the advertising effect.

[0034] Furthermore, the computer vision sentiment analysis model includes a visual feature extraction unit, a sentiment state reasoning unit, and a user experience score calculation unit;

[0035] The visual feature extraction unit is used to extract multi-dimensional features from the input facial image data, including facial expression features, eye movement trajectory data and sight stop information; the visual feature extraction unit detects facial key points through a convolutional neural network to obtain the user's eyebrows, eyes, and mouth corners. In combination with the optical flow method, the eye movement trajectory and sight movement direction of the user are calculated; the user's sight point is spatially mapped by using a deep learning model combined with the viewing angle parameters of the camera device, thereby generating accurate sight focus data and calibrating the user's gaze area in the advertising picture;

[0036] The emotional state inference unit is used to synthesize the output data of the visual feature extraction unit to infer the user's real-time emotional state, including interest, concentration and emotional tendency; the emotional state inference unit adopts a time series modeling method based on a long short-term memory network to identify the dynamic changes of the user's emotional state, and combines the self-attention mechanism to optimize the weight allocation of different visual features to improve the accuracy of emotional analysis; by jointly analyzing the change pattern of facial expressions, the stability of eye movement trajectories, the duration and area of ​​gaze retention, the user's interest in the current advertising content is calculated, the user's concentration level is evaluated, and the user's overall emotional tendency is identified;

[0037] The user experience score calculation unit is used to calculate the final user experience score based on the output result of the emotional state reasoning unit, and optimize the adaptability of the advertising content in combination with the line of sight focus data; the user experience score calculation unit adopts a weighted multi-factor scoring model to comprehensively consider the interest, concentration, emotional tendency and the stability of the user's line of sight focus to generate a user experience score, and makes personalized adjustments in combination with historical viewing behavior and individual preferences.

[0038] In this embodiment, the computer vision sentiment analysis model consists of a visual feature extraction unit, a sentiment state reasoning unit, and a user experience score calculation unit. These units cooperate with each other to extract and analyze the user's sentiment state from the input facial image data, and finally generate key data for dynamic optimization of advertising. The input and output of each unit are interrelated to ensure the consistency and effectiveness of the entire analysis process.

[0039] The main function of the visual feature extraction unit is to extract the key visual features of the user from the input facial image data, including facial expression features, eye movement trajectory data, and gaze retention information. The unit first uses a convolutional neural network to detect the key points of the user's face and identify the tiny movements of the eyebrows, eyes, corners of the mouth, etc. to extract the expression feature vector. Then, the user's eye movement trajectory is calculated in combination with the optical flow method to analyze the direction of his or her gaze movement, the change of the gaze area, and the gaze jump pattern. In addition, the unit uses a deep learning model combined with the viewing angle parameters of the camera device to spatially map the user's gaze point, generate accurate gaze focus data, and calibrate the user's specific gaze area in the advertising screen. The output of this unit includes expression feature vectors, eye movement trajectory data, and gaze focus data, which will be passed as input to the emotional state inference unit for more in-depth emotional analysis.

[0040] After receiving the output of the visual feature extraction unit, the emotional state inference unit infers the user's real-time emotional state based on these data, including interest, concentration and emotional tendency. The unit adopts a time series modeling method based on long short-term memory network, combined with the user's expression feature vector, to analyze the user's expression change trend at different time points to determine the dynamic evolution of emotions. At the same time, the unit uses the self-attention mechanism to optimize the weight distribution of different visual features, so that the eye movement trajectory data is combined with the expression features to calculate the user's interest. The calculation of interest is based on the user's gaze dwell time, eye movement trajectory stability and the intensity of attention to a specific advertising area. If the user stays on an advertising element for a long time and the gaze jumps less, the interest is high. The calculation of concentration depends on the user's eye movement fixation and the amplitude of the gaze deviation in a short period of time. If the user's gaze is too scattered or the gaze point is frequently changed, the concentration is low. The recognition of emotional tendency is based on the time series analysis of facial expressions. The system determines whether the user's current emotion is positive, neutral or negative by identifying the user's micro-expression features, such as a slight upward corner of the mouth or a tightened eyebrow. The output of this unit includes interest, concentration and emotional tendency, which will be input into the user experience score calculation unit to further evaluate the user's overall experience of the current advertising content.

[0041] The user experience scoring unit receives the output data of the emotional state inference unit and combines it with the gaze focus data to calculate the user's final experience score, so as to optimize the adaptability of the advertisement content. This unit adopts a weighted multi-factor scoring model, which comprehensively considers interest, concentration, and emotional tendency to generate the user experience score. During the scoring calculation process, interest is given a higher weight to measure the user's actual attention to the advertisement content, concentration is used to evaluate whether the user can continuously maintain attention, and emotional tendency is used to judge the user's acceptance of the advertisement content. In addition, this unit uses gaze focus data analysis to analyze the distribution of the user's fixation areas, determine the advertisement elements that the user mainly focuses on, and adjust the importance weights of the advertisement content when calculating the score. To improve the personalized matching of the score, this unit combines the user's historical viewing behavior data to dynamically optimize the score result, so that it can reflect the user's long-term interest trend. Finally, the output of this unit includes the user experience score, which will be used as the core basis for adjusting the advertisement placement to ensure that the advertisement content can accurately match the user's real-time emotional state and improve the advertisement placement effect and the user's viewing experience.

[0042] Through the collaborative processing of the visual feature extraction unit, the emotional state inference unit, and the user experience scoring unit, the computer vision emotion analysis model can accurately analyze the user's emotional state from the facial image data, and optimize the presentation method of the advertisement content based on the gaze focus and emotional feedback, making the advertisement placement more intelligent and personalized.

[0043] The following is the schematic reference implementation code of the computer vision emotion analysis model, which includes the visual feature extraction unit, the emotional state inference unit, and the user experience scoring unit. This code uses common deep learning libraries TensorFlow and OpenCV for calculations.

[0044] import cv2

[0045] import numpy as np

[0046] import tensorflow as tf

[0047] from tensorflow.keras.models import Model

[0048] from tensorflow.keras.layers import LSTM, Dense, Input, Attention

[0049] import dlib

[0050] # Visual feature extraction unit

[0051] class VisualFeatureExtractor:

[0052] def __init__(self, face_model_path):

[0053] """Initialize the facial landmark detection model"""

[0054] self.detector = dlib.get_frontal_face_detector()

[0055] self.predictor = dlib.shape_predictor(face_model_path)

[0056] def extract_facial_features(self, image):

[0057] """

[0058] Extract facial expression features

[0059] :param image: Input image (in BGR format)

[0060] :return: Landmark coordinates

[0061] """

[0062] gray = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)

[0063] faces = self.detector(gray)

[0064] features = []

[0065] for face in faces:

[0066] landmarks = self.predictor(gray, face)

[0067] for i in range(68):

[0068] x, y = landmarks.part(i).x, landmarks.part(i).y

[0069] features.append((x, y))

[0070] return np.array(features)

[0071] def compute_optical_flow(self, prev_frame, curr_frame):

[0072] """

[0073] Calculate the eye movement trajectory and the direction of gaze movement (using optical flow method)

[0074] :param prev_frame: Previous frame

[0075] :param curr_frame: Current frame

[0076] :return: Eye movement trajectory vector

[0077] """

[0078] prev_gray = cv2.cvtColor(prev_frame, cv2.COLOR_BGR2GRAY)

[0079] curr_gray = cv2.cvtColor(curr_frame, cv2.COLOR_BGR2GRAY)

[0080] # Shi-Tomasi corner detection

[0081] features = cv2.goodFeaturesToTrack(prev_gray, maxCorners=100,qualityLevel=0.3, minDistance=7)

[0082] if features is not None:

[0083] next_pts, status, _ = cv2.calcOpticalFlowPyrLK(prev_gray,curr_gray, features, None)

[0084] flow_vectors = next_pts - features

[0085] return flow_vectors

[0086] return None

[0087] def compute_gaze_focus(self, landmarks):

[0088] """

[0089] Combine facial key points and camera view to calculate the gaze focus

[0090] :param landmarks: Facial key point coordinates

[0091] :return: Gaze focus coordinates (screen position)

[0092] """

[0093] eye_region = landmarks[36:48] # Eye region

[0094] gaze_x = np.mean(eye_region[:, 0])

[0095] gaze_y = np.mean(eye_region[:, 1])

[0096] return gaze_x, gaze_y

[0097] # Emotion state inference unit

[0098] class EmotionStateInference:

[0099] def __init__(self, lstm_units=64):

[0100] """Initialize the Long Short-Term Memory (LSTM) network for emotion state modeling"""

[0101] self.model = self.build_lstm_model(lstm_units)

[0102] def build_lstm_model(self, lstm_units):

[0103] """

[0104] Build an LSTM model for processing sequential emotion data

[0105] :param lstm_units: Number of LSTM hidden layer units

[0106] :return: LSTM model

[0107] """

[0108] input_data = Input(shape=(10, 3)) # 10 time steps, each with 3 input features (interest level, concentration level, emotional tendency)

[0109] lstm = LSTM(lstm_units, return_sequences=True)(input_data)

[0110] attention = Attention()([lstm, lstm])

[0111] dense = Dense(3, activation='softmax')(attention) # Output 3 emotional states

[0112] model = Model(inputs=input_data, outputs=dense)

[0113] model.compile(optimizer='adam', loss='categorical_crossentropy')

[0114] return model

[0115] def infer_emotion_state(self, facial_features, eye_movement, gaze_focus):

[0116] """

[0117] Calculate the user's interest level, concentration level, and emotional tendency

[0118] :param facial_features: Facial expression features

[0119] :param eye_movement: Eye movement trajectory data

[0120] :param gaze_focus: Gaze focus data

[0121] :return: Predicted emotional state

[0122] """

[0123] interest_score = np.linalg.norm(gaze_focus - np.mean(facial_features, axis = 0)) # Attention distance

[0124] attention_score = 1.0 / (1.0 + np.linalg.norm(eye_movement)) # Eye movement stability

[0125] emotion_score = np.tanh(np.sum(facial_features)) # Emotional tendency (based on facial expression features)

[0126] input_data = np.array([[interest_score, attention_score, emotion_score]]) 10).reshape(1, 10, 3)

[0127] return self.model.predict(input_data)[0]

[0128] # User experience scoring calculation unit

[0129] class UserExperienceScoring:

[0130] def __init__(self):

[0131] """Initialize user experience scoring calculation"""

[0132] self.weights = {"interest": 0.5, "attention": 0.3, "emotion": 0.2} # Scoring weights

[0133] def compute_experience_score(self, emotion_state):

[0134] """

[0135] Calculate the final user experience score

[0136] :param emotion_state: Interest, concentration, and emotional tendency provided by the emotional state inference unit

[0137] :return: User experience score (between 0 and 1)

[0138] """

[0139] interest_score, attention_score, emotion_score = emotion_state

[0140] experience_score = (self.weights["interest"] interest_score +

[0141] self.weights["attention"] attention_score +

[0142] self.weights["emotion"] emotion_score)

[0143] return experience_score

[0144] # Computer vision emotion analysis model

[0145] class EmotionRecognitionModel:

[0146] def __init__(self, face_model_path):

[0147] """Initialize the computer vision emotion analysis model, including three units"""

[0148] self.visual_extractor = VisualFeatureExtractor(face_model_path)

[0149] self.emotion_inference = EmotionStateInference()

[0150] self.experience_scorer = UserExperienceScoring()

[0151] def analyze_emotion(self, prev_frame, curr_frame):

[0152] """

[0153] Run the entire analysis process to calculate the user's emotional state and experience score

[0154] :param prev_frame: Previous video frame

[0155] :param curr_frame: Current video frame

[0156] :return: User experience score

[0157] """

[0158] facial_features = self.visual_extractor.extract_facial_features(curr_frame)

[0159] eye_movement = self.visual_extractor.compute_optical_flow(prev_frame, curr_frame)

[0160] gaze_focus = self.visual_extractor.compute_gaze_focus(facial_features)

[0161] emotion_state = self.emotion_inference.infer_emotion_state(facial_features, eye_movement, gaze_focus)

[0162] experience_score = self.experience_scorer.compute_experience_score(emotion_state)

[0163] return experience_score

[0164] Training a computer vision emotion analysis model requires constructing a multi-modal dataset containing facial expressions, eye movement trajectories, and line-of-sight focus data. First, video data of users watching advertisements is collected, and facial key point detection algorithms are used to extract expression features. The optical flow method is used to calculate eye movement trajectories, and the line-of-sight landing points are mapped in combination with the perspective parameters of the camera device. Using these data as inputs, a convolutional neural network is trained to extract visual features, and a long short-term memory network is used to model the expression changes and eye movement patterns in the time series. To improve accuracy, a self-attention mechanism is introduced to optimize the weight allocation of visual features, enabling the model to focus on the most important information. During the training process, a weighted multi-factor scoring model is used to calculate the user experience score, the cross-entropy loss function is used to optimize the emotion classification task, and the mean square error is used to adjust the error of the user experience score prediction. Through multiple rounds of iterative training and combining individual preferences to adjust the model parameters, the accurate prediction of the user's emotional state is finally achieved, enabling the advertisement content to be optimized according to real-time emotional feedback.

[0165] Furthermore, the user experience score calculation unit uses the following formula 1 to calculate the user experience score:

[0166] ;

[0167] where, is the final user experience score; respectively represent the interest, concentration, and emotional tendency of the user at the th time step, and the value range is ; is the degree of deviation of the interest, concentration, and emotional tendency of the user at the th time step from their average level in the past specified time period; is the deviation of the interest, concentration, and emotional tendency of the user at the th time step from their average level in the past specified time period; is the time window size for calculating the user's historical emotional trend; is a small constant to prevent the denominator from approaching zero; is the emotion weight, which is used to dynamically adjust the influence of each time step in the overall score calculation, and is calculated using the following formula 2:

[0168] ;

[0169] where, are respectively the average interest, concentration, and emotional tendency of the user in watching advertisements in the past; is the personalized adjustment coefficient; is the time window size for calculating the user's historical emotional trend.

[0170] In this embodiment, during the calculation of the user experience score, a weighted scoring model that comprehensively considers the user's emotional changes, individual attention characteristics, and historical emotional trends is adopted to ensure that the advertisement content can accurately match the user's real-time emotional state, thereby optimizing the effect of advertisement placement. The user experience score is calculated by formula (1), and its core idea is to combine the emotional characteristics of the user at different time steps and assign dynamic weights to short-term emotional fluctuations to ensure the personalization and adaptability of the score.

[0171] The core variables of formula (1) include the user's interest, concentration, and emotional tendency at each time step, and these data are from the output of the emotional state inference unit. Interest represents the attractiveness of the current advertisement to the user, which is usually jointly determined by the user's gaze fixation time, facial expression changes, and eye movement stability. Its value range is between 0 and 1, where 1 represents extremely high interest and 0 represents complete lack of interest.

[0172] Concentration reflects the user's attention focus on the advertisement content. This indicator is calculated by combining the stability of the eye movement trajectory, the frequency of gaze jumps, and the user's blink rate. A higher concentration means that the user may be deeply understanding the advertisement content.

[0173] Emotional tendency is used to measure the user's overall emotional state, which is usually calculated through facial expression recognition and time series analysis. Its value range is also between 0 and 1, where 1 represents a positive emotion and 0 represents a negative emotion.

[0174] In formula (1), is the emotional weight of the user at the th time step, and its role is to dynamically adjust the contribution of this time step in the overall score calculation. The emotional weight is calculated by formula (2), and it is normalized and adjusted based on the deviation of the user's current emotional state from its historical average level. In formula (2), respectively represent the average interest, concentration, and emotional tendency of the user in the past time steps. They are used to measure the user's long-term emotional trend, so that short-term emotional fluctuations can be compared relative to the historical emotional state. By calculating the deviation between the user's current time step and its historical emotional trend and combining the exponential decay function to adjust the weight distribution, it is ensured that abnormal emotional fluctuations (such as sudden interest surges or loss of attention) can be reasonably reflected in the final score calculation.

[0175] In formula (1), the user's emotional variance Used to measure the amplitude of the user's emotional fluctuations at the current time step. It calculates the degree of deviation of interest, concentration, and emotional tendency from the historical emotional trend. A higher indicates that the user has significant emotional fluctuations at this time step, which may mean a sudden change in mood or that the advertisement has had a significant impact on the user.

[0176] Variance reflects the degree of deviation of the user's emotional state (interest, concentration, emotional tendency) at a certain time step from the user's overall emotional trend. It is used to measure the degree of fluctuation of the user's emotions, so that in the case of drastic changes in the user's emotions, the scoring calculation pays more attention to these moments, thereby enhancing the personalized adaptation ability.

[0177] The calculation method is as follows:

[0178] ;

[0179] Where respectively represent the interest, concentration, and emotional tendency of the user at the th time step; are respectively the average interest, concentration, and emotional tendency of the user in watching advertisements in the past period of time (such as the past time steps), and the calculation method is as follows:

[0180] ;

[0181] The variance of the emotional state at the time step represents the degree of deviation of the user's emotional value at the current time step from its historical average. When the user's emotional state at a certain time step significantly deviates from its historical trend (such as a sudden increase in interest or a drastic emotional fluctuation), the variance value at this time step will be larger, so that it occupies a higher weight in the final scoring calculation, improving the sensitivity to the user's current emotional state. If the user's emotions are relatively stable and the variance is small, the scoring calculation relies more on the long-term trend, making the personalized matching more stable.

[0182] In the scoring calculation, a ratio term of variance is introduced:

[0183] ;

[0184] Where represents the total emotional fluctuation amount in the past N time steps, and is a small constant to prevent the denominator from approaching zero, and the recommended value is . The role of this ratio is to ensure that in the case of large emotional fluctuations, the scoring calculation can pay more attention to short-term emotional changes, so as to adjust the adaptability of the advertisement content.

[0185] Personalized adjustment coefficient In formula (2), it is used to control the relative influence of interest, concentration, and emotional tendency in weight calculation. The recommended value is , , . A higher makes the change in interest have a greater impact on weight allocation, and is suitable for scenarios where the advertising content depends on user interest drive, such as entertainment or personalized recommendation ads. A higher makes the change in concentration have a stronger impact on weight allocation, and is suitable for scenarios where users need to actively understand the advertising content, such as financial products or technology product ads. A higher makes the influence of emotional tendency more obvious, and is suitable for emotion-driven ads, such as public welfare ads or brand story ads.

[0186] The time window size N represents the number of time steps used to calculate the user's historical emotional trend. The recommended value is generally between 10 and 30, specifically depending on the duration of the ad and the average viewing time of the user. A smaller N makes the score calculation more dependent on short-term emotional changes, while a larger N makes the system pay more attention to long-term emotional trends to ensure the stability and personalization of the score.

[0187] Finally, the user experience score is obtained by weighted summing the emotional data of all time steps and adjusting it in combination with emotional fluctuations to ensure that the score can comprehensively reflect the user's true experience of the ad. This calculation method enables the system to comprehensively consider long-term trends and short-term emotional fluctuations when evaluating the user's acceptance of the advertising content, and adapt to different user behavior patterns through personalized adjustment parameters, improving the accuracy of ad placement and the optimization effect of the user experience.

[0188] Step S103: Select the advertising content with the highest matching degree to the current user state from the preset advertising content library according to the user experience score.

[0189] In step S103, according to the user experience score, select the advertising content with the highest matching degree to the current user state from the preset advertising content library to ensure the accuracy and personalization of ad placement. First, the advertising content library needs to be constructed in advance and contain various types of advertising content. Each advertising content should have structured tag information, such as ad type, target audience, brand style, product category, color theme, context adaptability, etc. In addition, each advertising content can also contain historical placement data, user feedback data, and emotional matching weights to optimize the matching process.

[0190] To improve the accuracy of ad matching, the system first needs to parse the user experience score, which is composed of parameters such as interest level, attentiveness, and emotional tendency. The interest level reflects the user's potential interest in the current ad content, the attentiveness indicates the degree of the user's concentration while watching the ad, and the emotional tendency shows whether the user's emotional state is positive, neutral, or negative. Based on these metrics, a multi-dimensional vector can be defined to represent the current user's emotional state, and a comprehensive emotional score can be calculated by combining weights for use in ad matching.

[0191] The core of ad content matching lies in calculating the matching degree between the current user's emotional state and each ad content in the ad content library. The matching algorithm can be optimized using rule-based decision-making methods, machine learning models, or deep neural networks. For example, in the rule-based method, preset rules can be defined, such as "when the user's emotional tendency is positive and the interest level is high, give priority to pushing entertainment ads" or "when the user's attentiveness is low, select ad content with strong visual impact." In the machine learning method, collaborative filtering or content-based recommendation algorithms can be used to compare the user's current emotional characteristics with historical data to select the ad that best matches the user's state. In the deep learning method, a pre-trained emotion-ad matching model can be used, inputting the user's emotional state vector and outputting the best-matched ad content.

[0192] Once the ad content with the highest matching degree is calculated, the system will push the ad content according to the matching result. At the same time, to improve the real-time nature of the matching, the ad selection process can adopt a caching mechanism or index optimization to speed up the search and matching speed. If the ad content with the highest matching degree is the same as the current ad content, there is no need to switch. Otherwise, the system can trigger an update of the ad content and provide an optimization basis for subsequent steps. In addition, to improve the long-term matching accuracy, the system can store the user's viewing records, matching results, and user interaction feedback data, and use this data to continuously optimize the ad selection strategy to enhance the user experience and the commercial value of ad placement.

[0193] Furthermore, the step of selecting the ad content with the highest matching degree with the current user state from the preset ad content library according to the user experience score includes:

[0194] Based on the interest level, attentiveness, and emotional tendency in the user experience score, initially screen the ad content in the ad content library, eliminate ad types that do not match the current user's emotional state, and give priority to retaining ad content with positive feedback indicated by historical viewing data;

[0195] For the initially screened advertisement content, analyze the user's line-of-sight focus data, determine the characteristics of the advertisement elements that the user is more inclined to focus on during the current advertisement viewing process, including image areas, text information, or dynamic content, and calculate the matching degree of the candidate advertisements in the advertisement content library based on the distribution of the user's interest points, so as to improve the accuracy of advertisement matching;

[0196] When the matching degrees of multiple advertisement contents are close, combine the user's historical viewing behaviors, advertisement click records, and the preference data of users of the same type to adjust the recommended order of the advertisement contents.

[0197] When selecting the advertisement content with the highest matching degree with the current user status according to the user experience score, the system first needs to screen the advertisements in the advertisement content library to ensure that the recommended advertisement content can be consistent with the user's current emotional state, interest level, and concentration level. The interest degree, concentration degree, and emotional tendency in the user experience score jointly determine the screening criteria. Among them, the interest degree is used to judge the user's willingness to accept the advertisement content, the concentration degree is used to measure the user's current tolerance for information acquisition, and the emotional tendency reflects the user's potential preference for different types of advertisements. Based on these factors, the system screens the advertisement content library and eliminates those advertisement types that do not match the user's current emotional state. For example, when the user's emotional tendency is negative, the system will reduce the push of advertisements with high stimulation or high information density, and when the user's interest degree is high, it will preferentially select advertisements with detailed content and rich information. In addition, the system also refers to the user's historical viewing records and preferentially retains the data with longer past viewing time, more interaction behaviors, or positive feedback of the user to improve the accuracy of advertisement matching.

[0198] After the initial screening, the system further analyzes the user's line-of-sight focus data to determine which advertisement elements the user pays more attention to during the advertisement viewing process, including image areas, text information, or dynamic content. The system judges the user's interest degree in different advertisement elements through the user's fixation duration, eye movement trajectory, and line-of-sight fixation points, and calculates the matching degree between the candidate advertisement and the user's current interest points. For example, if the user's line of sight is more inclined to stay in the image area, the system will preferentially recommend advertisements with stronger visual impact and higher image proportion; if the user spends a longer time looking at the text information, it will select advertisement content with more intuitive and detailed information expression; for users who are concerned about dynamic content, it will recommend animation or short video advertisements to improve the attractiveness of the advertisement content. In this way, the system can ensure that the selected advertisement content is highly matched with the user's interest points, thereby enhancing the viewing experience and communication effect of the advertisement.

[0199] When the matching degrees of candidate advertisement contents are close, the system combines the user's historical viewing behaviors, advertisement click records, and preference data of users of the same type to optimize the sorting of advertisement recommendations. Historical viewing behaviors reflect the user's long-term interest trends and information acquisition habits. For example, if the user shows a high interest in a certain type of advertisement in multiple past viewings, the priority of this type of advertisement will be correspondingly increased. Advertisement click records are used to judge the actual interaction situation of the user with the advertisement content. For example, if the click-through rate of a certain type of advertisement is high, the recommendation weight of this type of advertisement will be increased. The preference data of users of the same type can be used to infer the user's potential interest points. If the system detects that users with similar behavior patterns to the current user have a high preference for a certain type of advertisement, the recommendation priority of this type of advertisement can be appropriately increased. Through this sorting optimization method based on multi-dimensional data, the system can select the advertisement that best meets the individual needs of the user from multiple advertisement contents with similar matching degrees, thereby further improving the accuracy of advertisement matching and the delivery effect.

[0200] Step S104: Dynamically adjust the visual layout and positions of key information elements of the selected advertisement content based on the line-of-sight focus data identified by the sentiment analysis model.

[0201] In step S104, the system dynamically adjusts the visual layout and positions of key information elements of the selected advertisement content based on the line-of-sight focus data identified by the sentiment analysis model to optimize the user's viewing experience and improve the effectiveness of the advertisement. First, it is necessary to analyze the line-of-sight focus data of the user, which includes but is not limited to the user's fixation area, line-of-sight dwell time, changing trend of eye movement trajectory, and position mapping relative to the screen content. The system uses this data to determine the user's attention area in the advertisement content, that is, the area where the user's line of sight is concentrated and the dwell time is long is usually the high-attention area, while the area where the line of sight moves quickly or stays briefly may indicate low attention or information redundancy.

[0202] After determining the user's attention area, the system optimizes and adjusts the layout of the advertisement content. Specifically, the system first analyzes the key information elements of the current advertisement, such as brand logo, product name, core selling points, price information, promotional content, and the main body of the image or video, and combines the user's line-of-sight focus data to rearrange the positions of these elements so that the key information can more easily enter the user's visual center. For example, if the user's line of sight is mainly concentrated in the left area of the advertisement screen and the core selling point of the advertisement is on the right, the core selling point information can be moved to the left to ensure that the user can quickly receive the key information. Similarly, if the user's line-of-sight dwell time is short or scattered, the system can focus the high-value information of the advertisement on the area where the line-of-sight dwells for a long time, thereby improving the effective transmission rate of the advertisement.

[0203] In terms of specific adjustment strategies, an adaptive layout optimization method based on regional division can be adopted. According to the grid structure of the advertisement screen, the advertisement content is divided into multiple dynamically adjustable regions. The system can adjust the position, size, or transparency of information blocks within these regions based on eye gaze data to enhance visual attractiveness. In addition, the system can combine an adaptive optimization algorithm based on reinforcement learning to learn the eye gaze distribution patterns of different user groups, so as to perform pre-optimization in subsequent advertisement displays, reduce adjustment latency, and improve the coherence of advertisement displays.

[0204] During the process of visual element adjustment, the system can also dynamically change the arrangement of the text and images in the advertisement. For example, for advertisement content with a long text description, if it is detected that the user's concentration is low, the amount of text can be reduced or the font size of keywords can be increased to make the key information more prominent. If the advertisement contains multiple products or services, the system can preferentially place the most relevant content in the user's high-concern area according to the user's interest level. In addition, if the advertisement is in video form, the system can optimize the display method of key frames based on the user's eye gaze focus, so that important information appears at the time point when the user's attention is highest, in order to improve the effective transmission rate of advertisement information.

[0205] When making layout adjustments, it is also necessary to consider visual balance and aesthetics to avoid abrupt changes in advertisement elements affecting the user experience. The system can adopt an image aesthetics evaluation model based on computer vision to ensure that the adjusted advertisement layout conforms to visual design principles, such as symmetry, contrast, and clarity of information hierarchy. For interactive advertisements, the system can also further optimize the layout in combination with the user's gestures or click behaviors. For example, when the user shows high concern for a certain area, the relevant content in that area can be enlarged or a button for further interaction can be provided to enhance user engagement.

[0206] The entire layout adjustment process can be carried out in real time during the advertisement display, or it can be pre-optimized based on historical user data to reduce the computational overhead and the impact of content mutations on the user experience. Finally, by dynamically adjusting the visual layout of the advertisement and the positions of key information elements, the advertisement content can more accurately adapt to the user's attention distribution, improving the dissemination effect of the advertisement and the user experience satisfaction.

[0207] Furthermore, dynamically adjusting the visual layout and the positions of key information elements of the selected advertisement content based on the eye gaze focus data identified by the computer vision emotion analysis model includes:

[0208] Based on the user's eye gaze focus data, determine the main attention area and low-attention area of the user in the advertisement content. Among them, when the user's eye gaze is concentrated on a specific area for a long time, this specific area is determined as the high-concern area, while the part where the user's eye gaze stays less is identified as the low-concern area;

[0209] For high - attention areas, optimize the display method of key information elements. Among them, if the advertisement content contains multiple information levels, the core brand logo, price information, or main advertisement copy should be preferentially adjusted to the high - attention area of users to improve the information transmission efficiency; weaken or reduce the non - core content in the low - attention area to reduce visual interference;

[0210] During the advertisement playback process, adaptively adjust dynamic advertisement elements in combination with the real - time changes of the user's line of sight. Among them, when it is detected that the user's line of sight stays in the key information area for a short time, appropriately enhance the visual prominence of this area, including enlarging the font, increasing the contrast, or adding animation effects to attract the user's attention. When the user pays long - term attention to the specified content, avoid excessive dynamic changes.

[0211] When adjusting the visual layout of advertisement content and the positions of key information elements based on the line - of - sight focus data recognized by the computer - vision emotion - analysis model, the system first needs to determine the main attention areas and low - attention areas of the user in the advertisement screen. Real - time collect the user's line - of - sight data during the user's viewing of the advertisement through a camera device, and conduct comprehensive analysis in combination with the line - of - sight stay time, eye movement trajectory, and focus stability to judge which advertisement areas the user stays in for a long time and which areas the line of sight only briefly sweeps across. The areas with long - term concentrated fixation are determined as high - attention areas, while the parts with less line - of - sight stay or frequent jumps are identified as low - attention areas. This area division can be dynamically adjusted according to the advertisement type and the user's visual behavior pattern to ensure that the system can accurately identify the user's actual focus points.

[0212] After completing the analysis of the user's attention areas, the system optimizes the display method of key information elements according to the high - attention areas to ensure that the most important information appears in the area that the user visually processes first. If the advertisement content contains multiple levels of information structures, such as brand logo, product introduction, price information, discount information, or advertisement copy, the system will preferentially adjust the core brand logo, price information, or content with strong marketing intent to the high - attention area of users to improve the information transmission efficiency and enable users to obtain key information faster. On the contrary, for the low - attention area, the system will appropriately weaken the visual weight of non - core content, such as reducing the font size, lowering the contrast, or reducing dynamic elements, to reduce visual interference, optimize the hierarchical relationship of advertisement information, make the overall advertisement structure more in line with the natural line - of - sight flow path of users, and improve the information absorption efficiency.

[0213] During the advertisement playback process, the system adaptively adjusts dynamic advertisement elements in combination with the real-time changes in the user's line of sight to ensure that the advertisement content can continuously attract the user's attention while avoiding information overload or interference. When it is detected that the user's line of sight stays in a key information area for a short time, the system will appropriately enhance the visual prominence of this area, such as enlarging the font, increasing the color contrast, or adding a slight animation effect, to strengthen the user's attention to this information. If the user stays in a certain key information area for a long time, the system will reduce the dynamic changes in this area to avoid causing user fatigue or distraction due to excessive visual stimulation. This adaptive adjustment strategy can be optimized according to the user's viewing pattern, so that the advertisement content reaches a balance between attracting attention and information stability, thereby improving the effectiveness of the advertisement and the user experience.

[0214] Step S105: Dynamically adjust the presentation mode of the advertisement content according to the real-time changes in the user experience score, including the content type, the picture-text ratio, and the information density.

[0215] In step S105, the system dynamically adjusts the presentation mode of the advertisement content according to the real-time changes in the user experience score to ensure that the advertisement content highly matches the user's emotional state and attention level, thereby improving the dissemination effect of the advertisement and the user's viewing experience. This adjustment process involves multiple aspects, including the optimization of the content type, the picture-text ratio, and the information density, to adapt to the user's current mood and concentration level.

[0216] First, in terms of the adjustment of the content type, the system will select the most suitable advertisement presentation mode for the user's current state according to the changes in the user experience score. When the user's interest score is relatively high and the concentration score is at a relatively high level, the system can choose to push more detailed advertisement content, such as video advertisements with in-depth product introductions or long-text advertisements, to maximize the transmission effect of advertisement information by taking advantage of the user's high attention. When the system detects that the user's interest is low or the concentration drops, more refined advertisement forms can be adopted, such as short video advertisements, dynamic picture advertisements, or short text content, to reduce the visual fatigue caused by excessive information for the user and improve the acceptance of information. In addition, if the user's emotional tendency is negative, the system can adjust the tone of the advertisement content and select advertisement materials that are more relaxed, have soft colors, or have positive emotional guidance to reduce the user's resistance and improve the affinity of the advertisement.

[0217] In terms of adjusting the picture - text ratio, the system will dynamically optimize the ratio of images to text in the advertisement based on the user's eye - focus data and content preferences. If the user shows a higher level of attention to the picture part of the advertisement and has a shorter reading time for the text content, the system can increase the proportion of the picture and reduce the density of the text information, making the advertisement content more intuitive and easy to understand, and conforming to the user's viewing habits. On the contrary, if the user's eye stays on the text area for a longer time, the system can increase the detail level of the text information, providing more product introductions or promotion details to fully meet the user's reading needs. The adjustment of the picture - text ratio is not limited to static advertisements but can also be applied to video advertisements. For example, it can adjust the size, display position of subtitles or highlight key information to adapt to different user preferences. In addition, the system can combine the user's historical behavior data to judge their preference trends for images and text and optimize the visual layout of the advertisement content accordingly to improve the efficiency of information transmission.

[0218] In terms of adjusting the information density, the system will adjust the content complexity of the advertisement according to the user's emotional state and changes in concentration. When the user is in a state of high concentration and shows continuous interest in the advertisement content, the system can increase the level of information details. For example, it can add more product specifications, usage cases or user reviews to the advertisement page to meet the user's need for in - depth understanding. When the user's concentration drops or their mood fluctuates greatly, the system will simplify the advertisement content, reduce the interference of irrelevant information, and highlight the most core advertisement points to make the information more concise and easy to read. For example, the system can automatically shorten the advertisement playback time, reduce the number of text lines on the page, or use the method of keyword highlighting so that users can obtain the core information of the advertisement in the shortest time. In addition, the system can combine the user's eye movement data to identify the range of information density that the user is most receptive to and adaptively optimize the advertisement according to this range to ensure the efficiency of information transmission.

[0219] During the entire dynamic adjustment process, the system not only relies on real - time user experience scores but also combines the user's historical data, industry standards, and advertisement effect evaluation models to continuously optimize the advertisement adjustment strategy. The system can adopt a machine - learning - based content recommendation algorithm to train the optimal advertisement presentation mode through historical data and continuously fine - tune it during real - time delivery to improve the personalized adaptation degree of the advertisement. In addition, to ensure the stability of the user experience, the system can adopt a progressive adjustment method instead of a mutation - type adjustment to reduce the visual discomfort caused by the rapid change of advertisement content, thus ensuring the natural transition of advertisement information. Finally, through the dynamic optimization of content type, picture - text ratio, and information density, it can effectively improve the matching degree of advertisement content, enhance the user's sense of participation, and the dissemination effect of the advertisement.

[0220] Furthermore, the dynamic adjustment of the advertisement content presentation mode according to the real - time change of the user experience score includes:

[0221] Based on the interest level, concentration level, and emotional tendency in the user experience score, determine the user's current information acceptance ability and preference characteristics, and set an advertising content adjustment strategy according to different score ranges. Among them, when the interest level and concentration level are relatively high, give priority to selecting advertising content with rich information. When the interest level or concentration level is relatively low, reduce non-core information to avoid users having a resistant attitude towards the advertisement.

[0222] For the adjusted advertising content, optimize the ratio of pictures to text. Among them, when the user's concentration level is relatively high and the interest level is relatively low, increase the proportion of image elements in the advertisement to enhance the visual impact. When the user's interest level is relatively high, increase the clarity and richness of text information to meet the user's in-depth reading needs.

[0223] During the advertisement playback process, continuously monitor the changing trend of the user experience score and dynamically adjust the information density. Among them, when the user's emotional tendency is positive and the concentration level is stable, increase the detailed presentation of the advertising content. When the user's concentration level drops or the emotional tendency tends to be negative, reduce the information density to improve the acceptability of the advertisement and the information transmission efficiency.

[0224] When dynamically adjusting the presentation method of advertising content according to the real-time changes in the user experience score, the system first comprehensively judges the user's current information acceptance ability and preference characteristics for advertising content based on the user's interest level, concentration level, and emotional tendency. The user experience score is used to measure the user's acceptance status of the current advertisement. The interest level reflects the attractiveness of the advertising content to the user. The concentration level indicates the degree of the user's attention focus. The emotional tendency determines the user's mental state when watching the advertisement. According to different combinations of these scores, the system formulates an adjustment strategy for advertising content. For example, when both the user's interest level and concentration level are at a relatively high level, the system gives priority to selecting advertising content with rich information and a complete structure to meet the user's in-depth understanding needs. When the user's interest level or concentration level is relatively low, the system reduces the non-core information in the advertisement to simplify the content presentation and avoid users having a resistant attitude or skipping the advertisement due to information overload.

[0225] After completing the preliminary adjustment of the advertisement content, the system further optimizes the graphic-text ratio to enhance the visual appeal of the advertisement and improve the information transmission effect. The relative proportion of images and text in the advertisement content is determined by the user's concentration and interest. When the user's concentration is high but the interest is low, the system increases the proportion of image elements in the advertisement and reduces long-text content to attract the user's attention with visual impact while ensuring that the information is presented intuitively and concisely. When the user's interest is high, the system enhances the clarity of the text information in the advertisement, increases the contrast and readability of the text, and provides more detailed product introductions or additional information to meet the user's in-depth reading needs. This adjustment method ensures that the expression form of the advertisement content conforms to the user's actual attention pattern and improves the effectiveness of information transmission.

[0226] During the advertisement playback process, the system continuously monitors the changing trend of the user experience score and dynamically adjusts the content information density of the advertisement accordingly. When the user's emotional tendency is positive and the concentration remains stable, the system gradually increases the detailed presentation of the advertisement content, such as adding more product descriptions, promotional information, or interactive elements to the screen to enhance the user's sense of participation and improve the conversion rate of the advertisement. When the user's concentration decreases or the emotion tends to be negative, the system reduces the information density, decreases excessive text content and visual elements to improve the acceptability of the advertisement and reduce the user's cognitive burden. This dynamic information density adjustment mechanism enables the advertisement to adapt to the user's emotional state and attention level, thereby optimizing the user experience and improving the dissemination effect and interaction rate of the advertisement.

[0227] Furthermore, the intelligent dynamic advertisement content placement method based on user emotion recognition further includes:

[0228] Set a visual heat detection area around the interface for displaying the advertisement content. The visual heat detection area does not display the actual advertisement content but can capture the user's peripheral visual attention.

[0229] Predict the user's visual fatigue degree and attention dispersion trend based on the user's gaze behavior in the visual heat detection area.

[0230] When it is detected that the user's visual fatigue exceeds the preset threshold, adjust the visual complexity and refresh frequency of the advertisement content to relieve the user's visual fatigue and optimize the advertisement experience.

[0231] In this embodiment, by setting a visual heat detection area around the interface for displaying advertisement content, the system can detect the peripheral visual attention of users, thereby providing an in-depth analysis of the overall visual state of users when watching advertisements. The visual heat detection area refers to a certain range around the advertisement interface. This area does not directly display advertisement content but is used to track the eye movement trajectories, gaze deviation patterns, and fixation conditions within the peripheral visual field of users. This detection area can adopt invisible sensing technology, enabling users to passively provide visual data without awareness, ensuring the naturalness and reliability of the detection data. Through a high-precision eye movement tracking algorithm, the system can capture information such as the fixation duration, gaze movement frequency, saccade behavior, and gaze deviation direction of users in this area.

[0232] Based on the gaze behavior of users collected in the visual heat detection area, the system predicts the degree of visual fatigue and the trend of attention dispersion of users. Visual fatigue can be judged by various parameters. For example, frequent fixation of users in the visual heat detection area may indicate a decrease in interest in the main advertisement content, while a long-term fixed fixation may imply a higher degree of visual fatigue of users. The system can also combine the blink frequency, pupil dilation, and nystagmus characteristics to comprehensively evaluate the visual comfort of users and determine whether there is a fatigue phenomenon caused by long-term high-stimulus visual input. In addition, the trend of attention dispersion can be analyzed through the frequency of users' gazes leaving the advertisement content, the increase in random eye movement trajectories, and the decrease in the fixation duration on the main advertisement elements. By constructing a real-time visual state model, the system can predict that users are about to enter the stage of attention decline or visual fatigue and accordingly dynamically adjust the advertisement presentation strategy to maintain the viewing experience of users.

[0233] When the system detects that the degree of visual fatigue of users exceeds a preset threshold, it will automatically adjust the visual complexity and refresh frequency of the advertisement content to relieve the visual fatigue of users and optimize the viewing experience of advertisements. The adjustment of visual complexity can be carried out by reducing the image detail density of the advertisement, decreasing the color saturation, lowering the contrast, or simplifying the visual elements, making the advertisement interface softer and reducing the visual burden on users. In addition, for dynamic advertisements, the system can reduce the speed of animation switching and reduce the visual stimulation of frequent flashing, thereby relieving the visual fatigue caused by rapid changes. The adjustment of the refresh frequency mainly targets the update rhythm of the advertisement. In the case of an increase in users' visual fatigue, the system can reduce the frequency of advertisement content switching, extend the display duration of the advertisement, or adopt a gradual transition method for content update to reduce abrupt visual changes and avoid interfering with users.

[0234] To further optimize the user experience, the system can also combine a personalized visual fatigue threshold adjustment strategy to dynamically adjust the control parameters of visual load according to the user's viewing history and preferences. For example, for users who watch ads for a long time, the system can set a lower visual fatigue trigger threshold to enter the low-stimulus mode earlier, while for users who watch ads with high concentration for a short time, the system can appropriately increase the visual complexity to ensure the effective conveyance of ad content. Ultimately, this method can achieve accurate prediction of users' visual fatigue and attention dispersion through intelligent monitoring of the visual heat detection area, and improve the long-term acceptability of ads and the quality of user experience by intelligently regulating the visual presentation of ad content.

[0235] Furthermore, the dynamic ad content intelligent delivery method based on user emotion recognition further includes:

[0236] Analyze the correlation between the user's facial micro-expressions and the gaze shift rate in real time to identify the subconscious interest points and resistance points of the user to the ad content;

[0237] Map the recognition results to the semantic partitions of the ad content to generate a user interest heat map;

[0238] Based on the user interest heat map, reconstruct the ad content in real time, strengthen the semantic elements that the user focuses on, and weaken or replace the content elements that trigger resistance emotions to achieve imperceptible personalized optimization of the ad content.

[0239] In this embodiment, by analyzing the correlation between the user's facial micro-expressions and the gaze shift rate in real time, the subconscious interest points and resistance points of the user to the ad content are identified. Facial micro-expressions are short-term and unconscious expression changes that can usually reflect the user's true emotional state. For example, a brief upward curl of the corners of the mouth may represent slight interest or pleasure, while a rapid downward pressure of the eyebrows may indicate dissatisfaction or boredom. To capture these micro-expression changes, the system uses a high-frame-rate camera and a deep learning expression recognition model to extract the feature data of expression changes by analyzing the user's facial muscle movement patterns. At the same time, combined with the user's gaze shift rate, that is, the speed and dwell time of the user's gaze movement between different regions of the ad content, the system can further analyze the user's active focus points and possible resistance areas of the ad content. When the user's gaze stays on an ad element for a long time and is accompanied by slight positive micro-expression changes, it can be judged that this area is a potential interest point of the user. If the user frequently and rapidly moves the gaze in a certain area or is accompanied by negative micro-expression changes, it may mean that the user has resistance to this content.

[0240] After obtaining the user's interest points and touch points, the system needs to map these recognition results to the semantic partitions of the advertisement content to generate a user interest heat map. Advertisement content usually contains multiple information elements, such as brand logos, product pictures, text descriptions, price information, discount labels, etc. These elements occupy different regions in the advertisement layout. To achieve semantic partition mapping, the system will perform structured processing on the advertisement content, divide different elements of the advertisement into regions, and classify the content semantics of each region based on natural language processing (NLP) or computer vision technology. For example, product descriptions are classified as informative content, brand logos are classified as brand recognition elements, and price discounts are classified as promotional stimulus elements. Then, the system matches the user's interest points and touch points with different semantic regions of the advertisement content to generate a user interest heat map, which can intuitively display the attention distribution of users to different parts of the advertisement. The interest heat map can be presented in a weighted color coding manner. Regions where users have a longer gaze time and a positive mood are marked as high-interest areas, while regions where users quickly glance over and are accompanied by negative emotions are marked as low-interest or touch-resistant areas.

[0241] After generating the user interest heat map, the system reconstructs the advertisement content in real time based on this data to optimize the advertisement display effect. The reconstruction methods of advertisement content include strengthening the semantic elements that users focus on, weakening the content elements that cause touch-resistant emotions, and even replacing the content in the touch-resistant areas in some cases to achieve seamless personalized optimization. Strengthening the semantic elements that users focus on can adopt dynamic adjustment methods. For example, the font size of the user's high-concern area can be enlarged, its contrast can be increased, or animation effects can be added to enhance the information transmission efficiency of this area. If the user has a strong interest in a certain product picture, the system can enhance the high definition of the picture area or extend its display duration so that users can understand this information more fully. For the content elements that cause the user's touch-resistant emotions, the system can reduce their visual weight. For example, reduce the color saturation, lower the brightness, reduce the dynamic effects, or adopt a fade-out process to reduce the interference to users without affecting the overall advertisement layout. In extreme cases, if a certain content element is determined to be a high-touch-resistant area, the system can use alternative materials in the advertisement content library for dynamic replacement. For example, if the user shows discomfort with a promotional label with overly bright colors, the system can replace it with a softer-toned version to improve the user's acceptance.

[0242] The entire optimization process takes place without the user's awareness, so it will not affect the overall coherence and viewing experience of the advertisement content. Since this method combines the user's real emotional feedback with the intelligent adjustment mechanism of the advertisement content, the advertisement can be dynamically adapted according to the real-time emotional state of individual users, thereby improving the user's acceptance of the advertisement, reducing advertisement interference, and ultimately enhancing the conversion effect of the advertisement. This unperceived personalized optimization method is more intelligent and efficient than the traditional static advertisement placement method, enabling the advertisement content to best suit the user's viewing preferences and emotional needs on the premise of ensuring the complete transmission of brand information.

[0243] Furthermore, the intelligent dynamic advertisement content placement method based on user emotion recognition further includes:

[0244] Monitoring the emotional fluctuations and attention change data of the user during the advertisement placement process;

[0245] When the emotional fluctuation data shows that the user's emotion tends to be negative or the attention change data shows that the user's attention decline exceeds a preset threshold, automatically trigger the buffer adjustment mode of the advertisement content;

[0246] In the buffer adjustment mode, based on the emotion guidance strategy, adjust the hue parameters, content rhythm speed, and background music characteristics of the advertisement, and switch the advertisement content from the high-stimulus information flow mode to the low-stimulus soothing mode;

[0247] Continuously monitor the user's emotional state, and when it is detected that the user's emotion index returns to the preset normal range, gradually resume the dynamic placement strategy of the original advertisement content.

[0248] In this embodiment, by monitoring the emotional fluctuations and attention change data of the user during the advertisement placement process, the intelligent adjustment of the advertisement content is realized, so that the advertisement placement can adapt to the user's real-time psychological state and optimize the viewing experience. The user's emotional fluctuation data is calculated based on facial expressions, eye movement trajectories, gaze fixation time, and other physiological indicators (such as blink frequency, facial muscle tension). The system uses a high-frame-rate camera device and a deep learning model to analyze the user's facial emotional changes in real time, and extracts features reflecting the emotional tendency, such as downturned corners of the mouth, frowning, or distracted eyes, which may indicate that the user has become bored, anxious, or has negative emotions. The attention change data is calculated based on the eye movement stability of the user, the frequency of gaze deviation, and the concentration degree of the gaze fixation area. When the user's gaze frequently moves away from the advertisement area, the fixation time shortens, or the eye movement pattern becomes irregular, it indicates that their attention is declining, and they may have become impatient or lost interest in the current advertisement content.

[0249] When it is detected that the user's emotion tends to be negative or the decrease in concentration exceeds the preset threshold, the system will automatically trigger the buffer adjustment mode of the advertisement content to reduce the visual or emotional burden on the user and prevent the user from skipping the advertisement or generating negative cognitions due to discomfort. The core of this buffer adjustment mode lies in dynamically adjusting the visual and auditory elements of the advertisement based on the emotion guidance strategy, switching the advertisement from a high-stimulus fast-paced mode to a softer and lower-stimulus soothing mode. Specifically, the adjustment of the hue parameter can make the picture softer by reducing the color saturation of the advertisement picture, reducing the visual impact of high-contrast areas, and increasing warm-tone elements, so as to reduce visual fatigue and the user's emotional fluctuations. The adjustment of the content rhythm speed can include reducing the animation switching rate in the advertisement, reducing the visual effect of rapid flickering, and extending the information display time, enabling the user to receive the advertisement content more comfortably. The adjustment of the background music characteristics includes reducing the volume, reducing high-frequency sound effects, and adopting a more soothing or low-tempo audio to reduce auditory stimulation, thereby creating a more peaceful advertisement atmosphere and preventing the user from having a resistant psychology due to excessive information stimulation.

[0250] In the buffer adjustment mode, the system will continuously monitor the user's emotional state to determine whether it has returned to the normal range. When it is detected that the user's emotion tends to be stable, the facial expression changes return to neutral or positive, the eye movement trajectory becomes more concentrated, and the concentration returns to a relatively high level, the system will gradually resume the dynamic delivery strategy of the original advertisement content. The recovery process adopts a progressive adjustment method to avoid sudden visual or auditory changes affecting the user experience. For example, the hue parameter can be slowly adjusted back to the original contrast, the animation rhythm can be gradually accelerated, and the volume and rhythm of the background music can also be adjusted according to a specific gradient curve to ensure that the user's emotion adapts to the change of the advertisement rhythm. To improve the overall user experience, the system can also automatically adjust the trigger threshold of the buffer adjustment mode in combination with the user's viewing history data and personalized emotion preferences to adapt to different users' emotional response patterns, making the advertisement delivery more personalized and intelligent. Through this method, the advertisement system can intelligently adjust the presentation mode of the advertisement content when the user's emotion fluctuates greatly or the attention decreases, thereby improving the user's acceptance of the advertisement, reducing interference, and at the same time enhancing the overall effectiveness of the advertisement delivery.

[0251] The second embodiment of this application provides an electronic device, which includes:

[0252] A processor;

[0253] A memory for storing a program, which when read and executed by the processor, executes a method for intelligent delivery of dynamic advertisement content based on user emotion recognition provided in the first embodiment of this application.

[0254] The third embodiment of the present application provides a computer-readable storage medium, on which a computer program is stored. When the program is executed by a processor, it executes a method for intelligent placement of dynamic advertisement content based on user emotion recognition provided in the first embodiment of the present application.

[0255] Although the present application is disclosed above in preferred embodiments, it is not intended to limit the present application. Any person skilled in the art can make possible changes and modifications without departing from the spirit and scope of the present application. Therefore, the protection scope of the present application should be subject to the scope defined by the claims of the present application.

Claims

1. A method for intelligent delivery of dynamic advertising content based on user emotion recognition, characterized in that: include: The facial image data of the user when watching the advertisement is collected in real time through a camera device, including facial expressions, eye movement trajectories, gaze lingering positions and lingering time; Inputting the facial image data into a computer vision sentiment analysis model to generate gaze focus data and user experience scores including interest, concentration, and emotional tendency; According to the user experience score, selecting the advertising content that best matches the current user status from a preset advertising content library; Dynamically adjust the visual layout and key information element positions of the selected advertising content based on the sight focus data identified by the computer vision sentiment analysis model; Based on the real-time changes in user experience scores, dynamically adjust the presentation of advertising content, including content type, image-text ratio, and information density.

2. The method for intelligent delivery of dynamic advertising content based on user emotion recognition according to claim 1, characterized in that: Also includes: Setting a visual heat detection area around the interface displaying the advertisement content, wherein the visual heat detection area does not display the actual advertisement content but can capture the user's marginal visual attention; Predicting the user's visual fatigue level and attention distraction tendency based on the user's gaze behavior in the visual heat detection area; When it is detected that the user's visual fatigue exceeds the preset threshold, the visual complexity and refresh frequency of the advertising content are adjusted to relieve the user's visual fatigue and optimize the advertising experience; The step of predicting the user's visual fatigue level and attention distraction tendency based on the user's gaze behavior in the visual heat detection area includes: Through high-precision eye tracking algorithms, the user's gaze dwell time, line of sight movement frequency, saccade behavior and line of sight deviation direction in the visual heat detection area are collected; Identify the user's frequent gaze behavior in the visual heat detection area as a signal of decreased interest in the main advertising content, or identify long-term fixed gaze as a signal of visual fatigue; The user's blinking frequency, pupil dilation and nystagmus characteristics are combined to comprehensively assess the degree of visual fatigue and attention distraction trends.

3. The method for intelligent delivery of dynamic advertising content based on user emotion recognition according to claim 1, characterized in that: Also includes: Real-time analysis of the correlation between the user's facial micro-expressions and the rate of eye movement, identifying the user's subconscious interest and resistance points to the advertising content; Map the recognition results with the semantic partitions of the advertising content to generate a user interest heat map; Based on the user interest heat map, the advertisement content is reconstructed in real time, the semantic elements that the user is concerned about are strengthened, and the content elements that cause resistance are weakened or replaced, thereby realizing the imperceptible personalized optimization of the advertisement content.

4. The method for intelligent delivery of dynamic advertising content based on user emotion recognition according to claim 1, characterized in that: Also includes: Monitor users’ emotional fluctuations and concentration changes during advertising; When the emotion fluctuation data shows that the user's emotion tends to be negative or the concentration change data shows that the decrease in the user's concentration exceeds a preset threshold, the buffer adjustment mode of the advertising content is automatically triggered; In the buffer adjustment mode, based on the emotion guidance strategy, the tonal parameters, content rhythm speed and background music characteristics of the advertisement are adjusted to switch the advertisement content from a high-stimulation information flow mode to a low-stimulation soothing mode; Continuously monitor the user's emotional state, and when it is detected that the user's emotional indicators return to the preset normal range, gradually restore the dynamic delivery strategy to the original advertising content.

5. The method for intelligent delivery of dynamic advertising content based on user emotion recognition according to claim 1, characterized in that: The computer vision sentiment analysis model includes a visual feature extraction unit, a sentiment state reasoning unit and a user experience score calculation unit; The visual feature extraction unit is used to extract multi-dimensional features from the input facial image data, including facial expression features, eye movement trajectory data and sight stop information; the visual feature extraction unit detects facial key points through a convolutional neural network to obtain the user's eyebrows, eyes, and mouth corners. In combination with the optical flow method, the user's eye movement trajectory and sight movement direction are calculated; in combination with the viewing angle parameters of the camera device, the user's sight point is spatially mapped to generate accurate sight focus data, and the user's gaze area in the advertising picture is calibrated; The emotional state inference unit is used to synthesize the output data of the visual feature extraction unit to infer the user's real-time emotional state, including interest, concentration and emotional tendency; the emotional state inference unit adopts a time series modeling method based on a long short-term memory network to identify the dynamic changes of the user's emotional state, and combines the self-attention mechanism to optimize the weight allocation of different visual features to improve the accuracy of emotional analysis; by jointly analyzing the change pattern of facial expressions, the stability of eye movement trajectories, the duration and area of ​​gaze retention, the user's interest in the current advertising content is calculated, the user's concentration level is evaluated, and the user's overall emotional tendency is identified; The user experience score calculation unit is used to calculate the final user experience score based on the output result of the emotional state reasoning unit, and optimize the adaptability of the advertising content in combination with the line of sight focus data; the user experience score calculation unit adopts a weighted multi-factor scoring model to comprehensively consider the interest, concentration, emotional tendency and the stability of the user's line of sight focus to generate a user experience score, and makes personalized adjustments in combination with historical viewing behavior and individual preferences.

6. The method for intelligent delivery of dynamic advertising content based on user emotion recognition according to claim 5, characterized in that: The user experience score calculation unit uses the following formula 1 to calculate the user experience score: ； in, Score the end-user experience; Respectively represent the user in The interest, concentration and emotional tendency of each time step ranges from ; For users in The degree of deviation of interest, concentration and emotional tendency at a time step from the average level in the past specified time period; For users in The degree of deviation of interest, concentration and emotional tendency at a time step from the average level in the past specified time period; is the time window size used to calculate the user's historical sentiment trend; is a tiny constant that prevents the denominator from going to zero; is the sentiment weight, which is used to dynamically adjust the influence of each time step in the overall score calculation, and is calculated using the following formula 2: ； in, They are the average interest, concentration and emotional tendency of the user in watching ads in the past; is the personalized adjustment coefficient; is the time window size used to calculate the user's historical sentiment trend.

7. The method for intelligent delivery of dynamic advertising content based on user emotion recognition according to claim 1, characterized in that: The real-time collection of facial image data of the user when watching the advertisement by means of a camera device includes: After detecting that the user's face enters the advertisement viewing area, the exposure parameters of the camera device are dynamically adjusted based on active light adaptation imaging technology to adapt to different ambient lighting conditions; Combined with an adaptive frame rate control algorithm, the sampling frame rate is increased when the user's eye movement frequency is high to accurately capture the instantaneous attention shift, and the sampling frequency is reduced when the user's gaze remains stable to reduce data redundancy and optimize computing efficiency.

8. The method for intelligent delivery of dynamic advertising content based on user emotion recognition according to claim 1, characterized in that: The selecting, according to the user experience score, the advertisement content with the highest matching degree with the current user status from a preset advertisement content library includes: Based on the interest, concentration and emotional tendency in the user experience score, the advertising content in the advertising content library is preliminarily screened to eliminate the types of advertisements that do not match the user's current emotional state, and to give priority to retaining the advertising content for which historical viewing data indicates that the user has positive feedback; For the initially screened advertising content, analyze the user's eye focus data to determine the characteristics of the advertising elements that the user is more likely to focus on during the current advertising viewing process, including image areas, text information or dynamic content, and calculate the matching degree of candidate advertisements in the advertising content library based on the distribution of the user's interest points to improve the accuracy of advertising matching; When the matching degrees of multiple advertising contents are close, the recommendation ranking of the advertising contents is adjusted based on the user's historical viewing behavior, ad click records and preference data of similar users.

9. The method for intelligent delivery of dynamic advertising content based on user emotion recognition according to claim 1, characterized in that: The method of dynamically adjusting the visual layout and key information element positions of the selected advertising content based on the sight focus data identified by the computer vision sentiment analysis model includes: Based on the user's sight focus data, determine the user's high attention area and low attention area in the advertising content, wherein when the user's sight is focused on a specific area for a long time, the specific area is determined as the high attention area, and the part where the user's sight is less focused is identified as the low attention area; Optimize the display of key information elements in high-attention areas. If the ad content contains multiple information levels, prioritize the core brand logo, price information or main ad copy to the high-attention area to improve information delivery efficiency. Weaken or reduce non-core content in low-attention areas to reduce visual interference. During the advertising playback process, dynamic advertising elements are adaptively adjusted in combination with the real-time changes in the user's line of sight. When it is detected that the user's line of sight stays in the key information area for a short time, the visual prominence of the area is appropriately enhanced, including enlarging the font, increasing the contrast, or adding animation effects to attract the user's attention. When the user pays attention to the specified content for a long time, excessive dynamic changes are avoided.

10. The method for intelligent delivery of dynamic advertising content based on user emotion recognition according to claim 1, characterized in that: The method of dynamically adjusting the presentation mode of the advertisement content according to the real-time changes of the user experience score includes: Based on the interest, concentration and emotional tendency in the user experience score, the user's current ability to accept information and preference characteristics are judged, and the advertising content adjustment strategy is set according to different score ranges. When the interest and concentration are high, the advertising content with rich information is given priority, and when the interest or concentration is low, the non-core information is reduced to avoid the user's resistance to the advertisement; Optimize the image-text ratio for the adjusted ad content. When the user's concentration is high and interest is low, increase the proportion of image elements in the ad to enhance the visual impact. When the user's interest is high, increase the clarity and content of the text information to meet the user's in-depth reading needs. During the advertising process, the changing trend of the user experience score is continuously monitored, and the information density is dynamically adjusted. When the user's emotional tendency is positive and the concentration is stable, the details of the advertising content are increased. When the user's concentration decreases or the emotion tends to be negative, the information density is reduced to improve the acceptability of the advertisement and the efficiency of information transmission.

Images