Brand promotion method and system based on user scene positioning

Abstract

The invention discloses a brand promotion method and system based on user scene positioning, and relates to the technical field of brand promotion, and the method comprises the specific steps: S100, data collection and integration: employing multiple types of intelligent sensors to form a sensing network, collecting the data of a physical environment where a user is located and the data of real-time position information, by applying various intelligent sensors and a data processing technology, the physical environment and position information of the user can be captured in real time, and the online behavior data of the user can be deeply mined, so that the accurate recognition of the user scene is realized, the pertinence of brand promotion is improved, and the user experience is improved. The promotion content can be closer to the actual demand and life scene of the user, the participation degree and acceptability of the user are improved through a personalized push mode, and compared with a traditional brand promotion mode, the precision and personalization are improved, and higher exposure rate and conversion rate are brought to the brand.

Description

Technical Field

[0001] The present invention relates to the technical field of brand promotion, and in particular to a brand promotion method and system based on user scenario positioning. Background Art

[0002] At a time when digital marketing is booming, brand promotion, as a key means for companies to enhance their market competitiveness and influence, is gradually developing in the direction of intelligence and personalization. With the continuous maturity of the Internet of Things, big data, and artificial intelligence technologies, brand promotion methods based on user scenario positioning have emerged, bringing new possibilities for brand promotion. This method uses an intelligent sensor network to collect real-time data on the user's physical environment and location information, and integrates the user's behavioral data on online platforms, providing a solid data foundation for accurately pushing brand content. The rapid development of data processing and analysis technology makes it possible to capture real-time scenes and predict future behaviors, thereby improving the pertinence and effectiveness of brand promotion.

[0003] Traditional brand promotion methods mostly rely on extensive advertising and standardized marketing information, ignoring the personalized needs and preferences of users in different scenarios. This single-mode promotion strategy is not only difficult to accurately reach target users, but may also cause user disgust and reduce brand reputation. In addition, traditional methods lack in-depth mining and analysis of user data, and cannot accurately predict future user behavior trends, thus making it impossible to formulate more forward-looking promotion strategies. At the same time, in terms of effect evaluation, traditional methods often rely on simple click-through rate and conversion rate indicators, ignoring the comprehensive consideration of multi-dimensional data such as user residence time and interaction rate, resulting in difficulty in comprehensive and accurate evaluation of promotion effects.

[0004] Therefore, developing a brand promotion method and system based on user scenario positioning will improve the efficiency and effectiveness of brand promotion and create greater commercial value for enterprises. Summary of the invention

[0005] The purpose of the present invention is to make up for the shortcomings of the prior art and to provide a brand promotion method and system based on user scenario positioning. Through intelligent sensor networks, data processing and integration, real-time scene capture, behavior scenario prediction, brand promotion content decision and push, and effect evaluation and feedback optimization steps, it achieves in-depth insight and accurate grasp of the user's physical environment and behavior patterns, and can push personalized brand promotion content based on the user's real-time scenario and future behavior prediction, thereby improving the accuracy and efficiency of brand communication.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions: On the one hand, a brand promotion method based on user scenario positioning, the specific steps of the promotion method are:

[0007] S100, Data Collection and Integration: Use various types of smart sensors to form a perception network to collect data related to the user's physical environment data and real-time location information, and each sensor transmits the collected data to the data processing center through a wireless communication network; at the same time, develop a customized data interface to extract detailed user purchase records from mainstream online shopping platforms, aggregate user interaction behavior data from social media platforms, extract user search history from search engines, collect user usage logs from applications, use ETL tools to convert and clean the collected data, and store the integrated data in a data warehouse;

[0008] S200, real-time scene capture: after receiving the data from the smart sensor, the data processing center comprehensively analyzes various environmental data and location information according to preset scene determination rules to determine the actual physical scene where the user is currently located;

[0009] S300, behavior scenario prediction: clean, denoise and convert the stored user behavior data, extract the user's behavior features through algorithms, build a prediction model through the exponential decay formula to predict the behavior scenarios that will be entered in the future, train the model using the user's historical behavior data, and predict user behavior through the trained model;

[0010] S400, brand promotion content decision and push: Based on the results of real-time scene capture and behavior scene prediction, combined with the brand promotion goals, the promotion content is determined through the promotion content decision formula, and pushed in the form of pictures, texts, videos, and pop-up windows using various applications on the user's smart terminal devices;

[0011] S500, effect evaluation and feedback optimization: Build a brand promotion effect evaluation system that includes user click-through rate, conversion rate, dwell time, and interaction rate indicators, receive user behavior data after promotion content, and analyze it. Based on the analysis results, optimize and adjust the promotion content format, copywriting, product matching, and scenario prediction accuracy.

[0012] Furthermore, in the S100, the intelligent sensors used in data collection and integration include temperature sensors, humidity sensors, light sensors, noise sensors, position sensors, and human flow sensors.

[0013] Furthermore, the preset scene determination rule in the instant scene capture in S200 is:

[0014] Outdoor leisure scene determination: When the value fed back by the temperature sensor is between 15℃-25℃, the humidity is between 40%-60%, the light intensity is between 1000lux-5000lux, the sound detected by the noise sensor is mainly natural ambient sound, and the location information shows that the user is in a park, garden, or suburban natural scenic area, it is determined that the user is in an outdoor leisure scene;

[0015] Commercial activity scene determination: When the temperature is between 20℃-30℃, the humidity is between 40%-70%, the light intensity is higher than 5000lux, the noise sensor captures a lot of noisy human voices, vehicle driving sounds and various commercial activity sounds, the noise decibel level is continuously higher than 70dB, the location information shows that the user is in the city's commercial street, shopping center surroundings, and market area, and the pedestrian flow sensor detects that there is more than one person per square meter per minute in the area, the user is determined to be in a commercial activity scene;

[0016] Traffic travel scene determination: When the location information shows that the user is at a railway station, bus station, airport, or subway station, and the temperature fluctuates between 10°C and 30°C, the humidity is between 30% and 70%, the light intensity is unstable, the noise decibel is high, and the sound sources are diverse, the user's moving speed is determined to be consistent with the normal moving speed of pedestrians in a transportation hub, which is 0.5-1.5 meters per second, and the user is determined to be in a traffic travel scene through positioning information;

[0017] Sports scene determination: During the day, the light intensity exceeds 5000 lux and is continuously stable, the temperature is between 18℃-30℃, the humidity is between 40%-60%, and the location information is displayed in the stadium, gymnasium, and school playground. At the same time, the noise sensor detects the regular sound of sports equipment being used, shouting, and the friction between sports shoes and the ground, and determines that the user is in an outdoor sports scene;

[0018] Home scene determination: The location information shows that the user is in a residential area, and the temperature is between 18℃-25℃, the humidity is between 30%-60%, the light intensity is less than 101lux at night and between 100lux-1000lux during the day, and the noise sensor detects a relatively quiet ambient sound, determining that the user is in a home scene.

[0019] Furthermore, in the above S300, the user behavior sequence feature vector is extracted by using the scene sequence behavior feature formula in the behavior scene prediction, and the formula is: Among them: F BS is the behavior sequence feature vector, L is the length of the behavior sequence, S i is the scene where the ith behavior occurs, d(S i-1 , S i ) are two adjacent scenes S i-1 and S iThe Euclidean distance between them, λ is the distance decay coefficient, H(S i ) is the behavior i in scene S i The heat value under i is the weight coefficient of behavior i.

[0020] Furthermore, in the behavior scenario prediction, the spatial behavior density clustering algorithm is used to extract geographic location features, and the geographic space is divided into a number of grid units. For each grid unit g, the user's stay time T in the unit is counted. g 、Visit times V g , the geographic location feature value H of the grid unit is calculated by the spatial behavior density clustering algorithm g , the formula is: Where β is the distance decay coefficient, D g It is the distance to surrounding commercial and entertainment venues.

[0021] Furthermore, in the behavior scenario prediction of S300, the probability of the user entering each potential future behavior scenario is calculated by a multi-factor future scenario probability formula, and the formula is: Where: P(S future ) represents the probability distribution of the user entering the future behavior scenario, S now is the scene the user is currently in, S i is the ith potential future behavior scenario, n is the total number of potential future behavior scenarios, d(S now , S i ) is a function that measures the similarity between the current scene and potential future scenes, F BS , H g 、F hist They are the user's current behavior sequence feature vector, geographic location feature vector, and historical behavior feature vector. is the historical average behavior sequence feature vector, geographic location feature vector, and historical behavior feature vector associated with the ith potential future behavior scenario, |·| represents the Euclidean distance between vectors, and ω i is the initial weight of each potential future behavior scenario, and λ, μ, and v are adjustment coefficients.

[0022] Furthermore, the promotion objectives in the S400, brand promotion content decision-making and push are: to enhance brand awareness, to increase brand exposure through extensive dissemination through online and offline channels; to enhance brand reputation, to shape a good image through high-quality products and services, public welfare activities and authoritative certification awards, and to win the trust of consumers; to promote product sales growth, to stimulate purchasing desire with limited-time discounts, combination packages, and gift strategies, and to increase market share and sales; to expand target market share, to identify new areas or groups through market research, to adjust positioning and strategies, and to expand brand coverage; to cultivate user loyalty, to maintain long-term relationships with consumers through high-quality customer service, personalized experience, and membership system benefits, and to promote word-of-mouth communication.

[0023] Furthermore, in the step S400, brand promotion content decision and push, promotion content is determined by a promotion content decision formula, and the set of instant scene capture factors is assumed to be S = {s1, s2, ..., s n}, the set of behavioral scenario prediction factors is B = {b1, b2, ..., b m}, the set of brand promotion target strategy factors is T = {t1, t2, ..., t k}, the set of brand promotion target strategy factors is T = {t1, t2, ..., t k}, the calculation formula of the comprehensive decision value P of the promotion content is: in are the corresponding weights of the instant scene capture factor set, the behavior scene prediction factor set, and the brand promotion target strategy factor set; n, m, and k are the corresponding numbers of factors in the instant scene capture factor set, the behavior scene prediction factor set, and the brand promotion target strategy factor set.

[0024] On the other hand, a brand promotion system based on user scenario positioning includes a data collection and integration module, an instant scenario capture module, a behavior scenario prediction module, a brand promotion content decision and push module, and an effect evaluation and feedback optimization module.

[0025] The data collection and integration module: utilizes a sensing network composed of temperature sensors, humidity sensors, light sensors, noise sensors, location sensors, and pedestrian flow sensors to collect the user's physical environment data and real-time location information, and transmits it to the data processing center through a wireless communication network; develops a customized data interface to extract detailed purchase records of users from online shopping platforms, aggregates user interaction data from social media platforms, extracts user search history from search engines, collects user usage logs from applications, and uses ETL tools to convert and clean the collected data, and stores the processed data in a data warehouse;

[0026] The real-time scene capture module: after receiving the intelligent sensor data, the data processing center comprehensively analyzes various environmental data and location information according to the preset scene determination rules to determine the actual physical scene where the user is currently located;

[0027] The behavior scenario prediction module: performs cleaning, denoising and format conversion operations on the stored user behavior data, extracts the user's behavior characteristics through an algorithm formula, constructs a prediction model through an exponential decay formula to predict the user's scenario in the future, and trains the model using the user's historical behavior data;

[0028] The brand promotion content decision and push module: determines the promotion content through the promotion content decision formula based on the instant scene capture and behavior scene prediction results, combined with the brand promotion goals, and pushes it through various applications on the user's smart terminal device;

[0029] The effect evaluation and feedback optimization module builds a brand promotion effect evaluation system that includes user click-through rate, conversion rate, dwell time, and interaction rate indicators, receives and analyzes user behavior data after the promotion content, and optimizes and adjusts the promotion content format, copywriting, product matching, and scenario prediction accuracy based on the analysis results.

[0030] Compared with the prior art, the brand promotion method and system based on user scenario positioning has the following beneficial effects:

[0031] 1. By comprehensively using a variety of intelligent sensors and data processing technologies, the present invention can capture the user's physical environment and location information in real time, and deeply mine the user's online behavior data, so as to realize accurate identification of user scenarios. It not only improves the pertinence of brand promotion, making the promotion content closer to the user's actual needs and life scenarios, but also improves the user's participation and acceptance through personalized push methods. Compared with traditional brand promotion methods, the present invention has achieved improvements in accuracy and personalization, bringing higher exposure and conversion rates to the brand.

[0032] 2. The present invention can construct a user behavior prediction model through deep learning and analysis of user historical behavior data, thereby accurately predicting the scenarios that users will enter in the future, so that brand promotion can be planned in advance and seize market opportunities. At the same time, the present invention also has the ability of dynamic optimization, and can adjust the promotion strategy and content in time according to the feedback of the promotion effect evaluation system to ensure the continuous effectiveness of the promotion activities. This mechanism of intelligent prediction and dynamic optimization not only improves the efficiency and effectiveness of brand promotion, but also brings stronger market competitiveness and adaptability to the brand.

[0033] Other advantages, objectives and features of the present invention will be set forth in part in the following description and, in part, will be apparent to those skilled in the art based on an examination of the following or may be taught from the practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention, and for ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0035] Figure 1 This is a process framework diagram of a brand promotion method based on user scenario positioning;

[0036] Figure 2 The flowchart of a brand promotion system based on user scenario positioning. DETAILED DESCRIPTION

[0037] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the specific implementation mode, structure, characteristics and effects of the present invention are described in detail below in combination with the accompanying drawings and preferred embodiments.

[0038] Embodiment 1

[0039] Brand promotion in outdoor leisure scenes

[0040] Data collection and integration, through the perception network composed of temperature sensors, humidity sensors, light sensors, noise sensors, location sensors and pedestrian flow sensors, data is continuously collected. For example, in the park, the temperature sensor detects that the temperature is 20℃, the humidity sensor shows that the humidity is 50%, the light sensor measures the light intensity as 3000lux, the noise sensor captures the natural ambient sound, the location sensor determines that the user is in the park, and the pedestrian flow sensor detects that the pedestrian flow in the park is moderate. These data are transmitted to the data processing center through the wireless communication network.

[0041] At the same time, we obtain records of users purchasing outdoor leisure products (such as tents and picnic mats) from online shopping platforms, collect data on users' interactive behavior of participating in outdoor topic discussions and liking outdoor photos from social media platforms, extract users' search history of "nearby parks" and "outdoor hiking routes" from search engines, and collect usage logs of users using map navigation to outdoor locations from applications, which are processed using ETL tools and stored in the data warehouse.

[0042] Instant scene capture: the data processing center comprehensively analyzes the collected data based on preset scene judgment rules. Since the temperature is between 15℃-25℃, the humidity is in the range of 40%-60%, the light intensity is between 1000lux-5000lux, the noise is natural ambient sound, and the location is in the park, it is determined that the user is in an outdoor leisure scene.

[0043] Behavior scenario prediction cleans, denoises, and converts the stored user behavior data, and extracts the user behavior sequence feature vector through the scenario sequence behavior feature formula. The formula is: Among them: F BS is the behavior sequence feature vector, L is the length of the behavior sequence, S i is the scene where the ith behavior occurs, d(S i-1 , S i ) are two adjacent scenes S i-1 and S i The Euclidean distance between them, λ is the distance decay coefficient, H(S i ) is the behavior i in scene S i The heat value under i is the weight coefficient of behavior i. The geographic location features are extracted through the spatial behavior density clustering algorithm, and the geographic space is divided into several grid cells. For each grid cell g, the user's stay time T in the cell is counted. g 、Visit times V g , the geographic location feature value H of the grid unit is calculated by the spatial behavior density clustering algorithm g , the formula is: Where β is the distance decay coefficient, D g is the distance to surrounding commercial and entertainment venues. For example, if users often engage in outdoor leisure activities on weekends, and most of the activities are held in surrounding parks, the probability of users entering each potential future behavior scenario is calculated using the multi-factor future scenario probability formula. The formula is: Where: P(S future ) represents the probability distribution of the user entering the future behavior scenario, S now is the scene the user is currently in, S i is the ith potential future behavior scenario, n is the total number of potential future behavior scenarios, d(S now , S i ) is a function that measures the similarity between the current scene and potential future scenes, F BS , H g 、F hist They are the user's current behavior sequence feature vector, geographic location feature vector, and historical behavior feature vector. is the historical average behavior sequence feature vector, geographic location feature vector, and historical behavior feature vector associated with the ith potential future behavior scenario, |·| represents the Euclidean distance between vectors, and ω i is the initial weight of each potential future behavior scenario, λ, μ, and v are adjustment coefficients. The model is trained using historical behavior data to predict that users may continue to visit nearby parks on weekends in the future.

[0044] Brand promotion content decision and push, combined with brand promotion goals (such as improving brand awareness, promoting outdoor product sales), determine the promotion content according to the promotion content decision formula, and set the instant scene capture factor set as S = {s1, s2, ..., s n}, the set of behavioral scenario prediction factors is B = {b1, b2, ..., b m}, the set of brand promotion target strategy factors is T = {t1, t2, ..., t k}, the set of brand promotion target strategy factors is T = {t1, t2, ..., t k}, the calculation formula of the comprehensive decision value P of the promotion content is: in are the corresponding weights of the instant scene capture factor set, the behavior scene prediction factor set, and the brand promotion target strategy factor set. n, m, and k are the corresponding numbers of factors in the instant scene capture factor set, the behavior scene prediction factor set, and the brand promotion target strategy factor set. For example, promotion information of a new tent of an outdoor brand is pushed to users, including the lightness and waterproofness of the tent, as well as limited-time discount activities, and is pushed in the form of pictures and texts through travel applications on users' smart terminal devices.

[0045] Effect evaluation and feedback optimization: The evaluation system constructed collects behavioral data of users after receiving promotional content, such as click-through rate and conversion rate. If it is found that the click-through rate of users on the pushed content is low, the analysis reason may be that the copy is not attractive enough or the product match is not high. In response to these problems, the copy is optimized, the product advantages are highlighted, the product recommendation strategy is adjusted, and the promotion effect is improved.

[0046] To sum up, this embodiment comprehensively collects data related to outdoor leisure scenes, accurately determines the scenes, estimates user behavior based on behavioral characteristics and scene prediction models, pushes adaptive promotion content according to decision formulas, and optimizes based on effect evaluation, so as to effectively improve the brand promotion effect in the outdoor leisure field, achieve brand promotion goals, and enhance user-brand interaction.

[0047] Embodiment 2:

[0048] Brand promotion in commercial activities

[0049] Data collection and integration, in the commercial street, the temperature sensor shows that the temperature is 25℃, the humidity sensor detects that the humidity is 55%, the light sensor measures the light intensity as 8000lux, the noise sensor captures noisy human voices, vehicle driving sounds and commercial activity sounds with the noise decibel continuously higher than 70dB, the location sensor determines that the user is located in the commercial street, and the traffic sensor detects that the traffic in this area is large (more than 1 person per square meter per minute), and transmits these sensor data to the data processing center.

[0050] Obtain users’ purchase records of fashion clothing and group-buying restaurants from online shopping platforms, gather users’ interactive data on their evaluations of stores in the shopping street and sharing of commercial activities from social media platforms, extract users’ search history of brands and commercial activity information in the shopping street from search engines, and collect users’ usage logs of payment applications and coupon applications in the shopping street from applications, and store them in the data warehouse after being processed by ETL tools.

[0051] Real-time scene capture: the data processing center comprehensively judges the temperature, humidity, light, noise, location and traffic information based on rules to determine whether the user is in a commercial activity scene.

[0052] Behavior scenario prediction, processing user behavior data, extracting the user behavior sequence feature vector through the scenario sequence behavior feature formula, which is: The spatial behavior density clustering algorithm is used to extract geographic location features and divide the geographic space into several grid cells. For each grid cell g, the user's stay time T in the cell is counted. g 、Visit times V g , the geographic location characteristic value H of the grid unit is calculated by the formula g , the formula is:

[0053] For example, if a user often goes shopping in the commercial street during holidays and prefers fashion brands, the probability of the user entering each potential future behavior scenario can be calculated using the multi-factor future scenario probability formula: The model is trained using historical user behavior data to predict that users are likely to go to the commercial street for fashion shopping during future holidays.

[0054] Brand promotion content decision and push, combined with brand promotion goals (such as improving brand reputation and promoting product sales), determine the promotion content through the promotion content decision formula, and set the instant scene capture factor set as S = {s1, s2, ..., s n}, the set of behavioral scenario prediction factors is B = {b1, b2, ..., b m}, the set of brand promotion target strategy factors is T = {t1, t2, ..., t k}, the set of brand promotion target strategy factors is T = {t1, t2, ..., t k}, the calculation formula of the comprehensive decision value P of the promotion content is: For example, information about the launch of new clothing from a fashion brand in a shopping street can be pushed to users, including the clothing's trendy designs and the high-quality fabrics used. At the same time, the discount promotions currently taking place in the store can be pushed. These information can be pushed in the form of pop-ups and videos through shopping applications on users' smart terminal devices. The videos show how models look wearing the new clothing.

[0055] Effect evaluation and feedback optimization: The evaluation system analyzes user behavior data. If the conversion rate is not high, it may be that the product display is not clear enough or the discount is not attractive enough. The promotion content format can be optimized, such as adding product detail pictures, adjusting discount strategies, and improving the accuracy of scene predictions, such as more accurately locating the user's preferred fashion style to improve the promotion effect.

[0056] To sum up, in commercial activity scenarios, this embodiment, after widely collecting data to determine the scenarios, uses behavioral feature analysis and scenario prediction methods, combines decision-making formulas to implement precise promotion, and improves based on effect feedback, successfully improving the brand's promotion efficiency in commercial scenarios, helping to improve brand reputation and promote product sales, and strengthening brand competitiveness.

[0057] The above description is only a preferred embodiment of the present invention and does not limit the present invention in any form. Although the present invention has been disclosed as a preferred embodiment as above, it is not used to limit the present invention. Any technical personnel in this field can make some changes or modify the technical contents disclosed above into equivalent embodiments without departing from the scope of the technical solution of the present invention. However, any brief modifications, equivalent changes and modifications made to the above embodiments based on the technical essence of the present invention without departing from the content of the technical solution of the present invention are still within the scope of the technical solution of the present invention.

Claims

1. A brand promotion method based on user scenario positioning, characterized in that: The specific steps of this promotion method are: S100, Data Collection and Integration: Use various types of smart sensors to form a perception network to collect data related to the user's physical environment data and real-time location information, and each sensor transmits the collected data to the data processing center through a wireless communication network; at the same time, develop a customized data interface to extract detailed user purchase records from mainstream online shopping platforms, aggregate user interaction behavior data from social media platforms, extract user search history from search engines, collect user usage logs from applications, use ETL tools to convert and clean the collected data, and store the integrated data in a data warehouse; S200, real-time scene capture: after receiving the data from the smart sensor, the data processing center comprehensively analyzes various environmental data and location information according to preset scene determination rules to determine the actual physical scene where the user is currently located; S300, behavior scenario prediction: clean, denoise and convert the stored user behavior data, extract the user's behavior features through algorithms, build a prediction model through the exponential decay formula to predict the behavior scenarios that will be entered in the future, train the model using the user's historical behavior data, and predict user behavior through the trained model; S400, brand promotion content decision and push: Based on the results of real-time scene capture and behavior scene prediction, combined with the brand promotion goals, the promotion content is determined through the promotion content decision formula, and pushed in the form of pictures, texts, videos, and pop-up windows using various applications on the user's smart terminal devices; S500, effect evaluation and feedback optimization: Build a brand promotion effect evaluation system that includes user click-through rate, conversion rate, dwell time, and interaction rate indicators, receive user behavior data after promotion content, and analyze it. Based on the analysis results, optimize and adjust the promotion content format, copywriting, product matching, and scenario prediction accuracy.

2. A brand promotion method based on user scenario positioning according to claim 1, characterized in that: The smart sensors used in the data collection and integration in S100 include temperature sensors, humidity sensors, light sensors, noise sensors, position sensors, and human flow sensors.

3. The brand promotion method based on user scenario positioning according to claim 1, characterized in that: S200, preset scene determination rules in real-time scene capture: Outdoor leisure scene determination: When the value fed back by the temperature sensor is between 15℃-25℃, the humidity is between 40%-60%, the light intensity is between 1000lux-5000lux, the sound detected by the noise sensor is mainly natural ambient sound, and the location information shows that the user is in a park, garden, or suburban natural scenic area, it is determined that the user is in an outdoor leisure scene; Commercial activity scene determination: When the temperature is between 20℃-30℃, the humidity is between 40%-70%, the light intensity is higher than 5000lux, the noise sensor captures a lot of noisy human voices, vehicle driving sounds and various commercial activity sounds, the noise decibel level is continuously higher than 70dB, the location information shows that the user is in the city's commercial street, shopping center surroundings, and market area, and the pedestrian flow sensor detects that there is more than one person per square meter per minute in the area, the user is determined to be in a commercial activity scene; Traffic travel scene determination: When the location information shows that the user is at a railway station, bus station, airport, or subway station, and the temperature fluctuates between 10°C and 30°C, the humidity is between 30% and 70%, the light intensity is unstable, the noise decibel is high, and the sound sources are diverse, the user's moving speed is determined to be consistent with the normal moving speed of pedestrians in a transportation hub, which is 0.5-1.5 meters per second, and the user is determined to be in a traffic travel scene through positioning information; Sports scene determination: During the day, the light intensity exceeds 5000 lux and is continuously stable, the temperature is between 18℃-30℃, the humidity is between 40%-60%, and the location information is displayed in the stadium, gymnasium, and school playground. At the same time, the noise sensor detects the regular sound of sports equipment being used, shouting, and the friction between sports shoes and the ground, and determines that the user is in an outdoor sports scene; Home scene determination: The location information shows that the user is in a residential area, and the temperature is between 18℃-25℃, the humidity is between 30%-60%, the light intensity is less than 10lux at night and between 100lux-1000lux during the day, and the noise sensor detects a relatively quiet ambient sound, determining that the user is in a home scene.

4. The brand promotion method based on user scenario positioning according to claim 1, characterized in that: In the above S300, in the behavior scenario prediction, the user behavior sequence feature vector is extracted by using the scenario sequence behavior feature formula, and the formula is: Among them: F BS is the behavior sequence feature vector, L is the length of the behavior sequence, S i is the scene where the ith behavior occurs, d(S i-1 , S i ) are two adjacent scenes S i-1 and S i The Euclidean distance between them, λ is the distance decay coefficient, H(S i ) is the behavior i in scene S i The heat value under i is the weight coefficient of behavior i.

5. The brand promotion method based on user scenario positioning according to claim 1, characterized in that: In the above S300, in the behavior scenario prediction, the geographic location features are extracted by using the spatial behavior density clustering algorithm, and the geographic space is divided into a number of grid units. For each grid unit g, the user's stay time T in the unit is counted. g 、Visit times V g , the geographic location feature value H of the grid unit is calculated by the spatial behavior density clustering algorithm g , the formula is: Where β is the distance decay coefficient, D g It is the distance to surrounding commercial and entertainment venues.

6. The brand promotion method based on user scenario positioning according to claim 1, characterized in that: In the above S300, the probability of the user entering each potential future behavior scenario is calculated by a multi-factor future scenario probability formula in the behavior scenario prediction, and the formula is: Where: P(S future ) represents the probability distribution of the user entering the future behavior scenario, S now is the scene the user is currently in, S i is the ith potential future behavior scenario, n is the total number of potential future behavior scenarios, d(S now , S i ) is a function that measures the similarity between the current scene and potential future scenes, F BS , H g 、F hist They are the user's current behavior sequence feature vector, geographic location feature vector, and historical behavior feature vector. is the historical average behavior sequence feature vector, geographic location feature vector, and historical behavior feature vector associated with the ith potential future behavior scenario, |·| represents the Euclidean distance between vectors, and ω i is the initial weight of each potential future behavior scenario, and λ, μ, and v are adjustment coefficients.

7. The brand promotion method based on user scenario positioning according to claim 1, characterized in that: The promotion objectives of S400 in brand promotion content decision-making and push are: to enhance brand awareness, to increase brand exposure through extensive dissemination through online and offline channels; to enhance brand reputation, to create a good image through high-quality products and services, public welfare activities and authoritative certification awards, and to win the trust of consumers; Promote product sales growth, stimulate purchasing desire with limited-time discounts, combination packages, and gift strategies, and increase market share and sales; expand target market share, identify new areas or groups through market research, adjust positioning and strategy, and expand brand coverage; cultivate user loyalty, maintain long-term relationships with consumers with high-quality customer service, personalized experience, and membership system benefits, and promote word-of-mouth communication.

8. The brand promotion method based on user scenario positioning according to claim 1, characterized in that: In S400, brand promotion content decision and push determine the promotion content through the promotion content decision formula, assuming that the real-time scene capture factor set is S = {s1, s2, ..., s n }, the set of behavioral scenario prediction factors is B = {b1, b2, ..., b m }, the set of brand promotion target strategy factors is T = {t1, t2, ..., t k }, the set of brand promotion target strategy factors is T = {t1, t2, ..., t k }, the calculation formula of the comprehensive decision value P of the promotion content is: in are the corresponding weights of the instant scene capture factor set, the behavior scene prediction factor set, and the brand promotion target strategy factor set; n, m, and k are the corresponding numbers of factors in the instant scene capture factor set, the behavior scene prediction factor set, and the brand promotion target strategy factor set.

9. A brand promotion system based on user scenario positioning, characterized in that: The system includes data collection and integration module, real-time scene capture module, behavior scene prediction module, brand promotion content decision and push module, and effect evaluation and feedback optimization module. The data collection and integration module: utilizes a sensing network composed of temperature sensors, humidity sensors, light sensors, noise sensors, location sensors, and pedestrian flow sensors to collect the user's physical environment data and real-time location information, and transmits it to the data processing center through a wireless communication network; develops a customized data interface to extract detailed purchase records of users from online shopping platforms, aggregates user interaction data from social media platforms, extracts user search history from search engines, collects user usage logs from applications, and uses ETL tools to convert and clean the collected data, and stores the processed data in a data warehouse; The real-time scene capture module: after receiving the intelligent sensor data, the data processing center comprehensively analyzes various environmental data and location information according to the preset scene determination rules to determine the actual physical scene where the user is currently located; The behavior scenario prediction module: performs cleaning, denoising and format conversion operations on the stored user behavior data, extracts the user's behavior characteristics through an algorithm formula, constructs a prediction model through an exponential decay formula to predict the user's scenario in the future, and trains the model using the user's historical behavior data; The brand promotion content decision and push module: determines the promotion content through the promotion content decision formula based on the instant scene capture and behavior scene prediction results, combined with the brand promotion goals, and pushes it through various applications on the user's smart terminal device; The effect evaluation and feedback optimization module builds a brand promotion effect evaluation system that includes user click-through rate, conversion rate, dwell time, and interaction rate indicators, receives and analyzes user behavior data after the promotion content, and optimizes and adjusts the promotion content format, copywriting, product matching, and scenario prediction accuracy based on the analysis results.

Images