Digital marketing method and system based on meta universe

Abstract

The invention relates to the technical field of digital marketing, in particular to a digital marketing method and system based on the meta universe, and the method comprises the following steps: based on user interaction data, extracting behavior features, calculating weights associated with the interaction frequency and page dwell time of a user on a meta universe platform, and carrying out feature clustering in combination with a purchase behavior, and predicting the periodic trend of the user activity, and generating a behavior characteristic index. According to the method, the interaction data of the user in the meta universe environment is analyzed, the time and the content of the marketing activity are accurately adjusted, the pertinence and the effect of the advertisement are remarkably improved, the periodic trend of the user behavior is predicted, the user response is monitored in real time, the marketing plan is enabled to quickly adapt to the change of the user group, and the user experience is improved. By comprehensively analyzing the character emotion and feedback of the user, the marketing plan is finely adjusted to match the preferences of different user groups, and the application of the A / B test allows the marketing plan to be quickly optimized based on the actual feedback, so that the conversion efficiency of the marketing activity is effectively enhanced.

Description

Technical Field

[0001] The present invention relates to the field of digital marketing technology, and in particular to a digital marketing method and system based on the metaverse. Background Art

[0002] Digital marketing involves the broad field of using digital technology to promote products or services, mainly through the Internet, social media, search engines, mobile devices and other digital channels to achieve marketing purposes, including but not limited to search engine optimization (SEO), content marketing, social media marketing, email marketing, affiliate marketing and data-based marketing automation strategies.

[0003] Among them, the digital marketing method of the metaverse is to implement marketing strategies in the virtual world or metaverse. The metaverse is a virtual environment constructed by augmented reality (AR), virtual reality (VR) and other advanced technologies. It provides a new platform where marketers can create immersive advertising experiences. Its purpose is to enhance users' stickiness to the brand and willingness to buy through virtual interaction and experience, while opening up a broader market and consumer groups.

[0004] Existing technologies rely too much on traditional platforms and static data analysis, lack real-time data processing capabilities, and are difficult to respond to changes in the market and user behavior. As a result, existing methods are inflexible in user behavior prediction and dynamic content adjustment, and cannot fully utilize rich user interaction data to optimize advertising effectiveness. Existing technologies fail to fully utilize the characteristics of the Metaverse platform, such as high user interactivity and immersive experience, making it impossible for marketing plans to effectively reach the target user groups, affecting the participation and effectiveness of advertising campaigns. These limitations are particularly evident in the rapidly developing Metaverse environment, because traditional digital marketing has failed to adapt to the needs of the emerging field of Metaverse in a timely manner. Summary of the invention

[0005] The purpose of the present invention is to solve the shortcomings existing in the prior art and to propose a digital marketing method and system based on the metaverse.

[0006] In order to achieve the above-mentioned purpose, the present invention adopts the following technical solution: a digital marketing method based on Metaverse, comprising the following steps: S1: Based on user interaction data, extract behavioral features, calculate the weights associated with the user's interaction frequency and page dwell time on the Metaverse platform, and perform feature clustering in combination with purchase behavior to predict the cyclical trend of user activities and generate behavioral feature indicators; S2: Based on the behavioral characteristic indicators, using historical data and prediction results, dynamically adjust the release time of marketing activities, and monitor the user's response data to the marketing activities in real time, adjust and optimize the marketing content that matches the user's preferences, and generate marketing response optimization configuration; S3: Based on the marketing response optimization configuration, analyze the keywords and sentiment indicators in the user feedback, match the preferences of different user groups in the metaverse, and use the A / B testing method to conduct multiple rounds of testing to generate test effect evaluation results; S4: Based on the test effect evaluation results, analyze the impact of differentiated marketing plans on user behavior, compare the effects of multiple marketing plans through conversion rate and user satisfaction data, monitor user reactions after the implementation of the marketing plans, and generate a marketing reaction overview.

[0007] The present invention is improved in that the calculation step of the associated weight is specifically as follows: S111: Based on the user interaction data, collect the user's interaction frequency and page dwell time data on the Metaverse platform, remove outliers from the data, and obtain processed user data; S112: Based on the processed user data, the correlation coefficient between the interaction frequency and the page dwell time is calculated using the formula: ; Generate the associated weights, where Represents the correlation weight between interaction frequency and page dwell time. Represents the page dwelling time of a single user. Represents the interaction frequency of a single user. and are the average page dwelling time and interaction frequency of all users respectively.

[0008] The present invention is improved in that the steps of obtaining the behavior characteristic index are specifically as follows: S121: combining the purchase behavior data, dividing the users into different groups through a clustering algorithm, and generating a feature clustering result; S122: Perform time series analysis on the feature clustering results to identify and quantify the behavioral changes of the groups in different time periods, using the formula: ; Predict future user behavior patterns and generate behavioral feature indicators ,in, Represents the time period, represents the adjustment factor of the amplitude, represents the frequency adjustment factor of the time period, represents the time offset, Indicates baseline adjustment.

[0009] The present invention is improved in that the step of dynamically adjusting the release time of the marketing activity is specifically as follows: S211: Based on the behavior characteristic indicators, extracting the user's response data to historical marketing activities, including click-through rate and response time, and removing invalid and abnormal data to obtain a response behavior data set; S212: Analyze the reaction behavior data set, and count the user activity levels in multiple time periods, using the formula: ; Calculation time period User activity score ,in, Represents the time period The number of active users in is the total number of users; S213: Based on the user activity score, identify the optimal marketing activity release period, adjust the marketing activity time, and obtain an optimized activity schedule.

[0010] The present invention is improved in that the step of obtaining the marketing response optimization configuration is specifically as follows: S221: monitoring the user's response data to the marketing activities in real time, including click-through rate, browsing time, and conversion rate indicators, collecting user behavior data, and obtaining real-time response data; S222: Analyze user preferences and behavior patterns based on the real-time response data, identify target needs and preferences of differentiated groups, and obtain a user preference information set; S223: Analyze the user preference information set and adjust the marketing content, including modifying the advertisement copy, design and promotion method to match the user's interests and needs and obtain the marketing response optimization configuration.

[0011] The present invention is improved in that the step of matching the preferences of the different user groups is specifically as follows: S311: Based on the marketing response optimization configuration, extract the comments and feedback information of users on the Metaverse platform on the marketing activities, and use natural language processing technology to extract keywords and sentiment indicators to obtain preliminary user feedback information; S312: Analyze the preliminary user feedback information using the formula: ; Calculate for each user group Preference rating , and obtain the difference user preference recognition results, where Representative keywords The frequency of occurrence, Represents sentiment indicators The score, and are the weight coefficients of keywords and sentiment indicators respectively.

[0012] The present invention is improved in that the steps of obtaining the test effect evaluation result are specifically as follows: S321: Using the A / B testing method, select a control group and multiple experimental groups, test the differential marketing plan on the Metaverse platform, record the response of each group of users to the marketing activities, and generate a test data set; S322: Analyze the test data set, use t-test to compare the performance differences between the control group and the experimental group, identify the effectiveness of various marketing plans, and evaluate their marketing effects to obtain test effect evaluation results.

[0013] The present invention is improved in that the step of obtaining the marketing response overview is specifically as follows: S411: Based on the test effect evaluation results, collect conversion rate and user satisfaction data from multiple marketing plans, including user click, purchase and satisfaction survey data, to obtain data preparation completion results; S412: Analyze the results of the data preparation, using the formula: ; Calculate the overall effectiveness of your marketing plan , and obtain the effect evaluation results, where Representative The conversion rate of each marketing plan, Indicates Customer satisfaction scores for each marketing plan, is the average of all marketing program conversion rates. is the average of the user satisfaction scores of all marketing programs, is the number of marketing plans; S413: According to the effect evaluation results, compare the effects of the marketing plans, evaluate the marketing plan with the best performance in terms of conversion rate and user satisfaction, and obtain an overview of the marketing response.

[0014] A digital marketing system based on the Metaverse, the system comprising: The interactive data analysis module collects user interaction data on the Metaverse platform, including page dwell time and interaction frequency, calculates association weights based on purchase behavior, and performs data clustering to predict the periodic trend of user activities and generate behavioral feature indicators; The marketing dynamic adjustment module dynamically adjusts the release time of marketing activities based on the behavioral characteristic indicators, using historical data and prediction results, and monitors the user's response data to the marketing activities in real time, adjusts and optimizes the marketing content that matches the user's preferences, and generates marketing response optimization configuration; The user feedback analysis module analyzes the keywords and sentiment indicators in the user feedback based on the marketing response optimization configuration, matches the preferences of different user groups in the metaverse, and uses the A / B testing method to conduct multiple rounds of testing to generate test effect evaluation results; The effect evaluation and optimization module analyzes the impact of the differential marketing plan on user behavior based on the test effect evaluation results, optimizes the marketing plan through conversion rate and user satisfaction data, and generates an optimized marketing process; Based on the optimized marketing process, the marketing monitoring module monitors the execution of marketing activities and real-time user feedback information, compares the effects of multiple marketing plans, monitors user reactions after the implementation of the marketing plans, and generates a marketing reaction overview.

[0015] Compared with the prior art, the advantages and positive effects of the present invention are: In the present invention, by analyzing the user's interactive data in the metaverse environment, the timing and content of marketing activities are accurately adjusted, and the pertinence and effectiveness of advertising are significantly improved. Not only is the cyclical trend of user behavior predicted, but also the real-time monitoring of user responses allows the marketing plan to quickly adapt to changes in user groups, greatly improving user experience and brand stickiness. By comprehensively analyzing the user's text emotions and feedback, the marketing plan is finely adjusted to match the preferences of different user groups. The application of A / B testing allows for rapid optimization of marketing plans based on actual feedback, effectively enhancing the conversion efficiency of marketing activities, and enabling effective interaction with consumers in the emerging field of the metaverse. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 A flowchart of a digital marketing method based on the Metaverse is proposed for the present invention; Figure 2 is a flow chart for calculating the associated weights in the present invention; Figure 3 This is a flow chart for obtaining behavior characteristic indicators in the present invention; Figure 4 A flowchart for dynamically adjusting the release time of a marketing activity in the present invention; Figure 5 A flowchart for obtaining the marketing response optimization configuration in the present invention; Figure 6 A flowchart for matching preferences of different user groups in the present invention; Figure 7 A flowchart for obtaining the test effect evaluation results in the present invention; Figure 8 The flowchart for obtaining the marketing response overview in the present invention. DETAILED DESCRIPTION

[0017] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0018] In the description of the present invention, it should be understood that the terms "length", "width", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", etc., indicating positions or positional relationships, are based on the positions or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, in the description of the present invention, the meaning of "multiple" is two or more, unless otherwise clearly and specifically defined.

[0019] Example: See Figure 1 The present invention provides a technical solution: a digital marketing method based on the metaverse, comprising the following steps: S1: Based on user interaction data, extract behavioral features, calculate the weights associated with the user's interaction frequency and page dwell time on the Metaverse platform, and perform feature clustering in combination with purchase behavior to predict the cyclical trend of user activities and generate behavioral feature indicators; S2: Based on behavioral characteristic indicators, using historical data and prediction results, dynamically adjust the release time of marketing activities, monitor user response data to marketing activities in real time, adjust and optimize marketing content that matches user preferences, and generate marketing response optimization configuration; S3: Based on the marketing response optimization configuration, analyze the keywords and sentiment indicators in the user feedback, match the preferences of different user groups in the metaverse, and use the A / B testing method to conduct multiple rounds of testing to generate test effect evaluation results; S4: Based on the test effect evaluation results, analyze the impact of differentiated marketing plans on user behavior, compare the effects of multiple marketing plans through conversion rate and user satisfaction data, monitor user reactions after the implementation of the marketing plans, and generate a marketing reaction overview.

[0020] Behavioral characteristic indicators include user interaction weight, time weight, and purchase tags. Marketing response optimization configuration specifically includes time adjustment parameters, content matching indicators, and preference adaptability metrics. Test effect evaluation results include test coverage, user feedback sensitivity, and plan adjustment effects. The marketing response overview specifically refers to satisfaction scores, conversion efficiency, and market acceptance.

[0021] See also Figure 2 , the calculation steps of the associated weight are as follows: S111: Based on the user interaction data, collect the user's interaction frequency and page dwell time data on the Metaverse platform, remove outliers from the data, and obtain processed user data; The interaction frequency and page dwell time data of Metaverse platform users are collected, and the outliers in all user data are processed. The interaction frequency data and page dwell time data are associated with each user. By comparing each user data point with the mean and standard deviation of the entire data, the data that deviates from the range of two times the standard deviation is eliminated, and the outliers are removed. A set of valid cross-sets of user interaction frequency and page dwell time data are obtained to generate processed user data.

[0022] S112: Based on the processed user data, the correlation coefficient between the interaction frequency and the page dwell time is calculated using the formula: ; Generate the associated weights, where Represents the correlation weight between interaction frequency and page dwell time. Represents the page dwelling time of a single user. Represents the interaction frequency of a single user. and They are the average page dwelling time and interaction frequency of all users respectively; Page dwell time ( ) and interaction frequency ( ) Collect and obtain user interaction data and calculate the average value of all users: ; A total of 100 user data were collected, of which one user's page dwell time was 200 seconds and the interaction frequency was 15 times. The total page dwell time and interaction frequency of other users were 18,000 seconds and 1,400 times respectively. , substitute the user data into the formula and calculate the numerator: ; Calculate the denominator: ; Substitute into the formula to calculate: ; The result shows that the correlation weight between page dwell time and interaction frequency on this user data is 1, indicating that the changes in the two are completely positively correlated.

[0023] See also Figure 3 , the specific steps for obtaining behavioral characteristic indicators are: S121: combining the purchase behavior data, dividing the users into different groups through a clustering algorithm, and generating feature clustering results; Collect and organize data on users' consumption habits, purchase frequency, and the amount of each purchase. Use numerical forms of user consumption records, such as using the average amount of each user's consumption as an indicator of consumption capacity, and using purchase frequency to calculate behavioral activity in the recent period. Combine the data with interactive behavior (such as page dwell time and interaction frequency) to form a user behavior feature matrix. Measure the similarity between users by calculating the Euclidean distance in the matrix, and perform grouping operations based on similarity to divide them into user groups with similar behavioral characteristics, so as to distinguish the behavioral patterns of different types of users and generate feature clustering results.

[0024] S122: Perform time series analysis on the feature clustering results to identify and quantify the behavioral changes of the groups under different time periods, using the formula: ; Predict future user behavior patterns and generate behavioral feature indicators ,in, Represents a time period in days, weeks, or months. The amplitude adjustment coefficient is used to amplify or reduce the amplitude of periodic fluctuations, reflecting the change in the intensity of user behavior. The frequency adjustment coefficient representing the time period is used to control the rate of periodic fluctuations and determines how fast the periodic behavior changes. Represents the time offset, which determines the specific time when the behavior trend cycle begins. It represents the baseline adjustment, which is a constant offset above or below the periodic fluctuation. It reflects the average level of user behavior and indicates the general state of user behavior even when there is no obvious periodic fluctuation. The behavior data of a user group shows periodic fluctuations. The data was collected for 30 days. Represents the time period, select sky, Adjust the amplitude, represents the daily frequency change, As the phase shift, represents the baseline adjustment, which can be calculated by substituting into the formula: ; ; ; ; ; The result shows that under the selected period and parameters, the behavioral characteristic index of the user group is 8.4028, which means that after periodic adjustment, the average behavioral characteristic index of the user group is above the medium level, reflecting its relatively high activity intensity and participation.

[0025] See also Figure 4 , the specific steps for dynamically adjusting the release time of marketing activities are as follows: S211: extracting user response data to historical marketing activities based on behavioral feature indicators, including click-through rate and response time, and removing invalid and abnormal data to obtain a response behavior data set; Extract users’ click-through rate and reaction time to historical marketing activities, organize the original data into time series form, record each user’s click event and corresponding timestamp respectively, remove duplicate records and unresponsive user data, filter the effective range of user reaction time by setting time difference threshold, delete reaction time data points that exceed the reasonable range, complete the removal of data outliers, and generate a reaction behavior data set including click-through rate and effective reaction time.

[0026] S212: Analyze the reaction behavior data set and count the user activity levels in multiple time periods using the formula: ; Calculation time period User activity score ,in, Represents the time period The number of active users in is the total number of users; 100 user activities are recorded between 10am and 11am, and the total number of users is 1000. Then: ; ; The result shows that the user activity during the time period from 10 to 11 o'clock is 0.1, or 10%, which helps the marketing team identify this time period as a potential optimized marketing campaign release period.

[0027] S213: Based on the user activity score, identify the optimal time period for releasing the marketing activity, ensure that the marketing activity is released when the user is most active, adjust the time of the marketing activity, and obtain an optimized activity schedule; Calculate the user activity level in multiple time periods, group the user behavior data by time period, such as hourly, half-hourly or other appropriate time intervals, count the number of active users in each time period, normalize the scores into active scores, and select the time period with the highest active score as the priority time window for releasing marketing activities by comparing the time period scores. Generate a time adjustment plan for the marketing activities based on the time period to form an optimized activity schedule.

[0028] See also Figure 5 , the specific steps for obtaining the marketing response optimization configuration are: S221: real-time monitoring of user response data to marketing activities, including click-through rate, browsing time and conversion rate indicators, collecting user behavior data, and obtaining real-time response data; Indicators are tracked continuously. Analytics and CRM are used in the data collection process to ensure data accuracy and timeliness. Illogical or obviously erroneous data points are identified and removed through preset thresholds and anomaly detection algorithms. For example, abnormally high click-through rates indicate robot interactions or system errors. This ensures that the collected data represents real and effective user behavior. Through this comprehensive monitoring and data screening mechanism, a reliable real-time response data set is obtained, which will be used for subsequent marketing strategy adjustments and content optimization work to ensure the effectiveness and pertinence of marketing activities.

[0029] S222: Analyze user preferences and behavior patterns based on real-time response data, identify target needs and preferences of differentiated groups, and obtain a user preference information set; By grouping and analyzing users' browsing behaviors and click records, statistics on users' behavioral characteristics and habitual preferences in different time periods, and extracting the behavioral patterns displayed by users in marketing activities, including users' click-through rates on specific types of marketing content (such as discount information or new product recommendations), and their response speed to push times, and by analyzing the cross-data of behavioral parameters and time characteristics, identifying the differentiated characteristics and demand preferences of user behavior patterns, and obtaining a user preference information set.

[0030] S223: Analyze the user preference information set and adjust the marketing content, including modifying the advertisement copy, design and promotion method, to match the user's interests and needs, and obtain the marketing response optimization configuration; Mapping the needs and preferences of user groups to specific marketing content improvements, including adjusting the theme expression of advertising copy according to user interests, optimizing the presentation order of promotional content, and restructuring marketing design based on user behavior history, ensuring that marketing content can highlight the core value points that the target user group is most concerned about. By associating copy modification records, visual design optimization suggestions and promotion form adjustment parameters, an optimized marketing response configuration plan is dynamically generated. This plan ensures that each content is directly related to the user's specific interests and behavior patterns, thereby achieving the goal of accurately matching user needs.

[0031] See also Figure 6 , the matching steps of preferences of different user groups are as follows: S311: Based on the marketing response optimization configuration, extract the user's comments and feedback on the marketing activities on the Metaverse platform, and use natural language processing technology to extract keywords and sentiment indicators to obtain preliminary user feedback information; Extract users’ comments and feedback on marketing activities on the Metaverse platform, obtain original text data generated by users through calls, including comments, likes, and other forms of interaction records, and then clean the data through text processing technology. The cleaning steps include removing meaningless characters, stop words, and correcting spelling errors. At the same time, uniformly translate multilingual content, decompose the cleaned text into independent terms through word segmentation technology, mark the decomposition results to identify high-frequency keywords, and combine sentiment analysis to evaluate the emotional tendencies shown by users in feedback, and generate preliminary user feedback information, which contains extracted keywords and corresponding sentiment indicators.

[0032] S312: Analyze the preliminary user feedback information using the formula: ; Calculate the number of users per group Preference rating , and obtain the difference user preference recognition results, where Representative keywords The frequency of occurrence, Represents sentiment indicators scores, measuring the intensity and frequency of emotions, and are the weight coefficients of keywords and sentiment indicators respectively; In a certain time window, the keyword "discount" of a user group appears 50 times, and its weight coefficient is set to 0.7. The score of the sentiment indicator "satisfaction" is 80 times, and the weight coefficient is set to 0.5. Substituting it into the formula, we get: ; ; ; The result shows that, based on the weighted calculation of keyword frequency and sentiment score, the preference score of this user group is 75, which reflects the overall positive response of users to the current marketing activities and provides data support for subsequent marketing strategies.

[0033] See also Figure 7 , the specific steps for obtaining the test effect evaluation results are: S321: Use the A / B testing method to select a control group and multiple experimental groups, test the differentiated marketing plan on the Metaverse platform, record the response of each group of users to the marketing activities, and generate a test data set; The marketing plan test is performed on the Metaverse platform. First, the groups are randomly assigned based on user behavior data to ensure uniformity between samples. Then, different marketing plan variables are set in the experimental group, such as content theme, push time, and activity reward mechanism. The default marketing strategy is maintained for the control group. The user behavior response data in each group is recorded in real time, including click-through rate, dwell time, and purchase conversion rate. A detailed response data set is formed through the records, thereby generating a test data set.

[0034] S322: Analyze the test data set, use t-test to compare the performance differences between the control group and the experimental group, identify the effectiveness of various marketing plans, and evaluate their marketing effects to obtain test effect evaluation results; The response data between the control group and the experimental group are compared through the t-test method. Response indicators such as click-through rate and purchase rate are set as key measurement parameters. The mean and variance of the experimental group on the indicators are calculated, and the difference analysis is performed with the control group data to screen out the experimental group that is statistically significantly better than the control group, identify the effectiveness of various marketing plans, summarize the differentiated performance into a marketing effectiveness report, and obtain the test effect evaluation results.

[0035] See also Figure 8 , the steps to obtain the marketing response overview are as follows: S411: Based on the test effect evaluation results, collect conversion rate and user satisfaction data from multiple marketing plans, including user click, purchase and satisfaction survey data, and obtain data preparation completion results; Conversion rate and user satisfaction data are collected from multiple marketing plans. User click behaviors of each marketing plan are classified and counted. The total number of user clicks and purchases for each plan are extracted. Satisfaction data is then obtained through user feedback questionnaires. The data is combined with marketing activity time and user behavior patterns, outliers and missing items are removed, and the data is uniformly formatted into a standardized table data set. The data is then labeled and organized by plan number for subsequent analysis to obtain the data preparation completion result.

[0036] S412: Prepare the data to complete the result analysis, using the formula: ; Calculate the overall effectiveness of your marketing plan , and obtain the effect evaluation results, where Representative The conversion rate of each marketing plan, Indicates Customer satisfaction scores for each marketing plan, is the average of all marketing program conversion rates. is the average of the user satisfaction scores of all marketing programs, is the number of marketing plans; There are 5 marketing plans with conversion rates of 10%, 15%, 12%, 8% and 20% respectively, and an average conversion rate of 13%. The user satisfaction scores are 80, 85, 78, 75 and 90 respectively, and the average satisfaction score is 81.6. The conversion rate is calculated as: ; ; Similarly, for user satisfaction: ; ; ; The results show that the overall effectiveness score of the marketing plan is 57.3, reflecting the comprehensive performance of the plan in reaching target users and meeting user needs.

[0037] S413: Based on the effect evaluation results, compare the effects of the marketing plans, evaluate the marketing plan with the best performance in terms of conversion rate and user satisfaction, and obtain an overview of the marketing response; First, we compare the conversion rate and user satisfaction data of each marketing plan, count the average conversion rate and satisfaction score of each plan, and calculate the deviation range of each plan. We then screen out outstanding plans by setting specific conversion rate and satisfaction thresholds. We then match the screening results with user behavior patterns and analyze them to identify the plan that best meets the needs of the target user group. We then mark it as the optimal plan and get a marketing response overview that describes the performance and differences of each plan.

[0038] Digital marketing system based on Metaverse, the system includes: The interactive data analysis module collects user interaction data on the Metaverse platform, including page dwell time and interaction frequency, calculates association weights based on purchase behavior, and performs data clustering to predict the periodic trend of user activities and generate behavioral feature indicators; The marketing dynamic adjustment module dynamically adjusts the release time of marketing activities based on behavioral characteristic indicators, using historical data and prediction results, and monitors user response data to marketing activities in real time, adjusts and optimizes marketing content that matches user preferences, and generates marketing response optimization configuration; The user feedback analysis module is based on the marketing response optimization configuration, analyzes the keywords and sentiment indicators in the user feedback, matches the preferences of different user groups in the metaverse, and uses the A / B testing method to conduct multiple rounds of testing to generate test effect evaluation results; The effect evaluation and optimization module analyzes the impact of differential marketing plans on user behavior based on the test effect evaluation results, optimizes the marketing plan through conversion rate and user satisfaction data, and generates an optimized marketing process; The marketing monitoring module monitors the execution of marketing activities and real-time user feedback based on the optimized marketing process, compares the effects of multiple marketing plans, monitors user reactions after the implementation of marketing plans, and generates a marketing reaction overview.

[0039] The above are only preferred embodiments of the present invention and are not intended to limit the present invention in other forms. Any technician familiar with the profession may use the technical contents disclosed above to change or modify them into equivalent embodiments with equivalent changes and apply them to other fields. However, any simple modification, equivalent change and modification made to the above embodiments based on the technical essence of the present invention without departing from the technical solution of the present invention still falls within the protection scope of the technical solution of the present invention.

Claims

1. A digital marketing method based on the Metaverse, characterized by: The following steps are involved: Based on user interaction data, we extract behavioral features, calculate the weights associated with the user's interaction frequency and page dwell time on the Metaverse platform, and perform feature clustering in combination with purchase behavior to predict the cyclical trend of user activities and generate behavioral feature indicators; Based on the behavioral characteristic indicators, using historical data and prediction results, dynamically adjust the release time of marketing activities, and monitor the user's response data to marketing activities in real time, adjust and optimize the marketing content that matches the user's preferences, and generate marketing response optimization configuration; Based on the marketing response optimization configuration, analyze the keywords and sentiment indicators in the user feedback, match the preferences of different user groups in the metaverse, and use the A / B testing method to conduct multiple rounds of testing to generate test effect evaluation results; Based on the test effect evaluation results, analyze the impact of differentiated marketing plans on user behavior, compare the effects of multiple marketing plans through conversion rate and user satisfaction data, monitor user reactions after the implementation of the marketing plans, and generate a marketing reaction overview.

2. The digital marketing method based on the Metaverse according to claim 1, characterized in that: The calculation steps of the associated weight are specifically as follows: Based on user interaction data, collect the user interaction frequency and page dwell time data on the Metaverse platform, remove outliers from the data, and obtain processed user data; Based on the processed user data, the correlation coefficient between the interaction frequency and the page dwell time is calculated using the formula: ； Generate the associated weights, where Represents the correlation weight between interaction frequency and page dwell time. Represents the page dwelling time of a single user. Represents the interaction frequency of a single user. and are the average page dwelling time and interaction frequency of all users respectively.

3. The digital marketing method based on the Metaverse according to claim 1, characterized in that: The steps for obtaining the behavior characteristic index are specifically as follows: In combination with the purchase behavior data, the users are divided into different groups through a clustering algorithm to generate feature clustering results; Perform time series analysis on the feature clustering results to identify and quantify the behavioral changes of the group under different time periods, using the formula: ； Predict future user behavior patterns and generate behavioral feature indicators ,in, Represents the time period, represents the adjustment factor of the amplitude, represents the frequency adjustment factor of the time period, represents the time offset, Indicates baseline adjustment.

4. The digital marketing method based on the Metaverse according to claim 1, characterized in that: The specific steps for dynamically adjusting the release time of the marketing activity are as follows: Based on the behavioral characteristic indicators, extract the user's response data to historical marketing activities, including click-through rate and response time, and remove invalid and abnormal data to obtain a response behavior data set; The reaction behavior data set is analyzed to count the user activity levels in multiple time periods, using the formula: ； Calculation time period User activity score ,in, Represents the time period The number of active users in is the total number of users; Based on the user activity score, the optimal marketing activity release period is identified, and the marketing activity time is adjusted to obtain an optimized activity schedule.

5. The digital marketing method based on the Metaverse according to claim 1, characterized in that: The steps for obtaining the marketing response optimization configuration are specifically as follows: Monitor the user's response data to the marketing activities in real time, including click-through rate, browsing time and conversion rate indicators, collect user behavior data, and obtain real-time response data; Based on the real-time response data, analyzing user preferences and behavior patterns, identifying target needs and preferences of differentiated groups, and obtaining a user preference information set; Analyze the user preference information set and adjust the marketing content, including modifying the advertising copy, design and promotion method, to match the user's interests and needs and obtain the marketing response optimization configuration.

6. The digital marketing method based on the Metaverse according to claim 1, characterized in that: The steps for matching the preferences of different user groups are specifically as follows: Based on the marketing response optimization configuration, extract the comments and feedback information of users on the Metaverse platform on the marketing activities, and use natural language processing technology to extract keywords and sentiment indicators to obtain preliminary user feedback information; The preliminary user feedback information is analyzed using the formula: ； Calculate for each user group Preference rating , and obtain the difference user preference recognition results, where Representative keywords The frequency of occurrence, Represents sentiment indicators The score, and are the weight coefficients of keywords and sentiment indicators respectively.

7. The digital marketing method based on the Metaverse according to claim 1, characterized in that: The steps for obtaining the test effect evaluation results are specifically as follows: Using the A / B testing method, select a control group and multiple experimental groups, test the differentiated marketing plan on the Metaverse platform, record the response of each group of users to the marketing activities, and generate a test data set; The test data set is analyzed, and the performance differences between the control group and the experimental group are compared using the t-test test to identify the effectiveness of various marketing plans and evaluate their marketing effects to obtain test effect evaluation results.

8. The digital marketing method based on the Metaverse according to claim 1, characterized in that: The steps for obtaining the marketing response overview are specifically as follows: Based on the test effect evaluation results, collect conversion rate and user satisfaction data from multiple marketing plans, including user click, purchase and satisfaction survey data, to obtain data preparation completion results; The data was prepared to complete the result analysis, using the formula: ； Calculate the overall effectiveness of your marketing plan , and obtain the effect evaluation results, where Representative The conversion rate of each marketing plan, Indicates Customer satisfaction scores for each marketing plan, is the average of all marketing program conversion rates. is the average of the user satisfaction scores of all marketing programs, is the number of marketing plans; Based on the effectiveness evaluation results, the effectiveness of marketing plans are compared, and the marketing plan with the best performance in terms of conversion rate and user satisfaction is evaluated to obtain an overview of marketing responses.

9. The digital marketing system based on Metaverse is characterized by: According to any one of claims 1 to 8, the digital marketing method based on the metaverse is implemented, and the system comprises: The interactive data analysis module collects user interaction data on the Metaverse platform, including page dwell time and interaction frequency, calculates association weights based on purchase behavior, and performs data clustering to predict the periodic trend of user activities and generate behavioral feature indicators; The marketing dynamic adjustment module dynamically adjusts the release time of marketing activities based on the behavioral characteristic indicators, using historical data and prediction results, and monitors the user's response data to the marketing activities in real time, adjusts and optimizes the marketing content that matches the user's preferences, and generates marketing response optimization configuration; The user feedback analysis module analyzes the keywords and sentiment indicators in the user feedback based on the marketing response optimization configuration, matches the preferences of different user groups in the metaverse, and uses the A / B testing method to conduct multiple rounds of testing to generate test effect evaluation results; The effect evaluation and optimization module analyzes the impact of the differential marketing plan on user behavior based on the test effect evaluation results, optimizes the marketing plan through conversion rate and user satisfaction data, and generates an optimized marketing process; Based on the optimized marketing process, the marketing monitoring module monitors the execution of marketing activities and real-time user feedback information, compares the effects of multiple marketing plans, monitors user reactions after the implementation of the marketing plans, and generates a marketing reaction overview.

Images