Intelligent sharing yoga lesson booking platform system

Abstract

The invention discloses an intelligent sharing yoga lesson booking platform system, and relates to the technical field of intelligent sharing, which comprises the following steps: a user registers an account and verifies the identity; analyzing social media activities, comments and feedback contents of the user, and identifying an emotional tendency of the user by using a natural language processing technology; personalized course recommendation is provided; evaluating the yoga skill level of the user, and planning an optimal learning path for the user; the user reserves a course according to the recommended course list and completes a payment process; feedback is submitted after the course ends, so that personalized course recommendation and learning path planning are improved; according to the method, highly personalized course recommendation is provided by using a natural language processing technology, so that the user satisfaction is improved; the skill level is evaluated and an optimal learning path is planned by applying a reinforcement learning algorithm, so that skill improvement is promoted; and the recommendation strategy and the learning path planning are continuously improved by collecting user feedback, so that the self-learning and adaptive capacity of the system is enhanced, and the service quality is further improved.

Description

Technical Field

[0001] The present invention relates to the field of intelligent sharing technology, and in particular to an intelligent sharing yoga class appointment platform system. Background Art

[0002] With the rapid development of information technology, intelligent yoga appointment platforms have become one of the important tools for people to pursue a healthy lifestyle. Traditional yoga appointment platforms often rely on simple user information collection and basic course recommendation algorithms, lacking personalized services and in-depth sentiment analysis. In recent years, yoga appointment platforms have gradually emerged, providing yoga enthusiasts with convenient appointment services and personalized course recommendations. However, traditional yoga appointment platforms often lack effective user identity authentication methods, making it difficult to ensure user information security; at the same time, most platforms fail to make full use of users' social media activities and feedback information when recommending courses, resulting in the recommended courses often not being personalized enough. In addition, existing yoga appointment platforms usually use traditional payment methods, lacking transparency and security. Summary of the invention

[0003] In view of the above existing problems, the present invention is proposed.

[0004] Therefore, the present invention provides an intelligent shared yoga class booking platform system to solve the problems of insecure identity authentication, non-personalized course recommendations, and opaque payment processes on existing yoga class booking platforms.

[0005] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0006] In a first aspect, an embodiment of the present invention provides an intelligent shared yoga class booking platform system, which includes a user registration and identity authentication module, which is responsible for the user to register an account on the yoga class booking platform and verify the identity through the zero-knowledge proof technology based on blockchain;

[0007] Sentiment analysis module, which is responsible for analyzing users’ social media activities, comments, and feedback, and using natural language processing technology to identify users’ sentiment tendencies;

[0008] The personalized course recommendation module is responsible for providing personalized course recommendations based on the recognition results of the user's emotional tendencies;

[0009] The learning path planning module is responsible for evaluating the user's yoga skill level and using reinforcement learning algorithms to plan the best learning path for the user;

[0010] The course reservation and payment module is responsible for users to reserve courses based on the recommended course list and complete the payment process through blockchain technology;

[0011] The feedback collection and optimization module is responsible for submitting feedback after the course ends to improve personalized course recommendations and learning path planning.

[0012] As a preferred solution of the intelligent shared yoga class booking platform system of the present invention, the user registers an account on the yoga class booking platform and verifies the identity through the zero-knowledge proof technology based on blockchain, including the following steps:

[0013] The user visits the yoga class booking platform, clicks the register button to enter the registration page, fills in the registration information, reads and agrees to the platform's terms of service and privacy policy, and clicks the submit button to complete the registration;

[0014] After the user completes the registration, he / she enters the identity verification page. The platform uses blockchain-based zero-knowledge proof technology to verify the user's identity information. After the verification is passed, the platform generates a unique blockchain identifier for the user, which is associated with the user's account.

[0015] If the verification fails, modify the incorrect user information and re-verify until the verification passes.

[0016] As a preferred solution of the intelligent shared yoga class booking platform system of the present invention, analyzing the content of the user's social media activities, comments and feedback includes the following steps:

[0017] Collect text data of users’ social media activities, comments, and feedback, split the text data into individual words, and remove non-text information;

[0018] The TF-IDF value of each word is calculated using TF-IDF technology to obtain text features, and the feature representation is further enhanced using word embedding technology.

[0019] As a preferred solution of the intelligent shared yoga class booking platform system of the present invention, the natural language processing technology is used to calculate the overall sentiment tendency score of the user, including the following steps:

[0020] Use the BERT model to perform sentiment analysis on each enhanced text feature, expressed as,

[0021] f(w) = N(w; θ);

[0022] Where N represents the BERT model, w represents the enhanced text features, θ represents the parameters of the model, and f(w) represents the sentiment category score;

[0023] Based on each user's historical emotional tendency, set the emotional category classification threshold;

[0024] When f(w) is greater than the threshold for emotional classification, the text sentiment is defined as positive; when f(w) is less than the threshold for emotional classification, the text sentiment is defined as negative; and when f(w) is equal to the threshold for emotional classification, the text sentiment is defined as neutral.

[0025] Define the corresponding sentiment score for each text sentiment tendency based on f(w);

[0026] The sentiment scores in all the user's texts are summed and aggregated, and the normalization function is applied to obtain the user's overall sentiment tendency score, which is expressed as,

[0027]

[0028] Where S(u) represents the overall sentiment score of user u, tanh represents the hyperbolic tangent function, and T u represents the set of all texts of user u, where u represents the user, c(w) represents the sentiment tendency of the text, and s c(w) represents the sentiment score corresponding to c(w), and β represents the adjustment parameter of the contribution of each text sentiment score.

[0029] As a preferred solution of the intelligent shared yoga class booking platform system of the present invention, the emotional tendency of the user is identified based on the overall emotional tendency score of the user, including the following steps:

[0030] Based on each user's historical emotional tendency, set the overall emotional category division threshold;

[0031] When S(u) is greater than the overall sentiment category division threshold, the user's sentiment tendency is defined as positive;

[0032] When S(u) is less than the overall emotion classification threshold, the user's emotion tendency is defined as negative;

[0033] When S(u) is equal to the overall emotion classification threshold, the user's emotion tendency is defined as neutral.

[0034] As a preferred solution of the intelligent shared yoga class booking platform system of the present invention, the personalized course recommendation is provided based on the recognition result of the user's emotional tendency, including the following steps:

[0035] In the interval where the user's overall sentiment tendency is positive, provide personalized courses to the user and calculate the recommendation score of each personalized course, which is expressed as,

[0036]

[0037] Among them, R c represents the personalized course recommendation score of course c, U represents the user set, and s crepresents the sentiment tendency score of course c, λ represents the parameter that adjusts the steepness of the function, and μ represents the average sentiment tendency score of all courses;

[0038] According to the calculated recommendation score of each personalized course, personalized courses are sorted in descending order, and the top k personalized courses are recommended to the user.

[0039] As a preferred solution of the intelligent shared yoga class booking platform system of the present invention, the following steps are included in the process of using a questionnaire survey to understand the user and evaluate the user's yoga skill level:

[0040] Design a questionnaire survey on the yoga class booking platform to fully understand user needs through various methods;

[0041] Based on the questionnaire data filled out by users, we design a skill level evaluation index, formulate an evaluation standard based on various aspects of yoga skills, and calculate the user's skill level score, which is expressed as,

[0042]

[0043] Among them, L u represents the skill level score of user u, n represents the number of evaluation indicators, x i represents the user's score on the i-th evaluation indicator, σ i The standard deviation of the i-th evaluation indicator, where i represents the number of evaluation indicators;

[0044] Set skill level indicators based on professional knowledge of yoga skills;

[0045] When L u When it is greater than the skill level indicator, the user's yoga skill level is defined as advanced;

[0046] When L u When the skill level is less than or equal to the skill level indicator, the user's yoga skill level is defined as elementary.

[0047] As a preferred solution of the intelligent shared yoga class booking platform system of the present invention, the following steps are included: based on the user's yoga skill level, a reinforcement learning algorithm is used to plan the best learning path for the user:

[0048] Define the state space and action space of the learning path, and design a reward function to encourage users to complete a higher quality learning path;

[0049] Use the DQN algorithm to plan the best learning path, define the state transition probability according to the user's state and the action taken, and use the ε-greedy strategy to balance exploration and utilization, expressed as,

[0050]

[0051] Among them, Q(s,a) represents the expected cumulative reward of taking action a in state s, a represents the learning rate, r represents the immediate reward, γ represents the discount factor, and s ′ Represents the new state after taking action a, a ′ Represents the new state s ′ The action below, θ - represents the parameters of the target network, and s represents the state;

[0052] When selecting an action, an action is randomly selected with probability ∈ for exploration, and the action with the largest expected cumulative reward in the current state is selected with probability 1-∈ to plan the best learning path for the user.

[0053] As a preferred solution of the intelligent shared yoga class booking platform system of the present invention, the user books a class according to the recommended course list and completes the payment process through blockchain technology, including the following steps:

[0054] Based on the user's personalized course recommendation and learning path planning, the most suitable yoga course list is recommended to the user. The user chooses the course he wants to book according to his needs and schedule. After the user confirms the booking information, the yoga booking platform generates a payment request. The user completes the payment verification through the blockchain wallet. After the payment is completed, the transaction record is added to the blockchain;

[0055] Use Solidity to write smart contracts in the payment process to ensure the smooth completion of the transaction.

[0056] As a preferred solution of the intelligent shared yoga class booking platform system of the present invention, wherein: submitting feedback after the course is over to improve personalized course recommendations and learning path planning includes the following steps:

[0057] After arriving at the yoga studio, users sign in on the yoga class booking platform’s app. After the class, users submit feedback on the yoga class booking platform.

[0058] Clean the feedback data submitted by users, remove invalid or duplicate data, use natural language processing technology to perform sentiment analysis on open feedback, quantify user satisfaction, calculate user feedback scores based on the feedback scoring formula, and adjust the weights of feedback dimensions based on the importance of the feedback content;

[0059] Add the new feedback data to the model's training dataset and retrain the deep reinforcement learning model using the updated dataset;

[0060] Based on the new model output, update the recommendation algorithm, adjust the recommendation strategy according to user feedback and changes in skill level, and update the learning path based on user feedback and new recommendation strategies.

[0061] The beneficial effects of the present invention are as follows: by verifying the identity of the user through the zero-knowledge proof technology based on blockchain, the user's privacy is protected and the trust of the platform is improved. The natural language processing technology is used to analyze the user's social media activities, comments and feedback content, identify emotional tendencies, and then provide highly personalized course recommendations, which greatly improves user satisfaction. By evaluating the user's yoga skill level and using the reinforcement learning algorithm to plan the best learning path, the user's skill improvement and personal progress are effectively promoted. The use of blockchain technology to complete the payment process not only ensures the security and transparency of the transaction, but also reduces the risk of fraud and improves payment efficiency. By collecting user feedback and continuously improving personalized course recommendations and learning path planning, the system's self-learning and adaptability are enhanced, and the service quality is further improved. The yoga class appointment platform has become a safe, reliable, highly personalized and constantly improving service platform. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0063] Figure 1 This is a flow chart of the intelligent shared yoga class booking platform system in Example 1.

[0064] Figure 2 This is a diagram showing the structure of the recommended course list in Example 1. DETAILED DESCRIPTION

[0065] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings.

[0066] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0067] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0068] Example 1, reference Figure 1 and Figure 2 , which is the first embodiment of the present invention, and provides an intelligent shared yoga class booking platform system, comprising the following steps:

[0069] S1. The user registers an account on the yoga class booking platform and verifies his identity through blockchain-based zero-knowledge proof technology.

[0070] The user visits the yoga class booking platform, clicks the register button to enter the registration page, fills in the registration information (including name, email address, create a password, etc.), reads and agrees to the platform's terms of service and privacy policy, and clicks the submit button to complete the registration;

[0071] After the user completes the registration, he / she enters the identity verification page. The platform uses blockchain-based zero-knowledge proof technology to verify the user's identity information. After the verification is passed, the platform generates a unique blockchain identifier for the user, which is associated with the user's account.

[0072] If the verification fails, modify the incorrect user information and re-verify until the verification passes.

[0073] It should be noted that users complete identity authentication through blockchain-based zero-knowledge proof technology, which ensures the security of user information while simplifying the registration process, improving user experience, increasing user data security, enhancing the trust of the platform, and reducing the risks caused by identity leakage.

[0074] S2. Analyze the content of users’ social media activities, comments, and feedback, and use natural language processing technology to identify users’ emotional tendencies.

[0075] Collect text data of users’ social media activities, comments, and feedback, split the text data into individual words, and remove non-text information (including stop words, punctuation, numbers, etc.);

[0076] The TF-IDF (term frequency-inverse document frequency) technology is used to calculate the TF-IDF value of each word. The TF-IDF value is equal to the product of the term frequency and the inverse document frequency to obtain text features. The term frequency is for each document in the document collection. The number of times each word appears is counted. The number of occurrences of each word is divided by the total number of words in the document to obtain the term frequency of the word in this document. The inverse document frequency counts the number of all documents containing a certain word. The total number of documents is divided by the number of documents containing the word, and then the logarithm is taken to obtain the inverse document frequency of the word. The feature representation is further enhanced through word embedding technology (such as Word2Vec or BERT). The specific operation is to use a pre-trained word embedding model to convert each word into a corresponding word embedding vector. These word embedding vectors can capture the similarity and contextual information between word meanings. The word embedding vectors of all words in each document are averaged to obtain the word embedding vector representation of the document, which enhances the feature representation.

[0077] The BERT (Bidirectional Encoder Representation) model is used to perform sentiment analysis on each enhanced text feature, expressed as,

[0078] f(w) = N(w; θ);

[0079] Where N represents the BERT model, w represents the enhanced text features, θ represents the parameters of the model (including weights and biases in the model), and f(w) represents the sentiment category score;

[0080] Based on each user's historical emotional tendency, set the emotional category classification threshold;

[0081] When f(w) is greater than the threshold for emotional classification, the text sentiment is defined as positive; when f(w) is less than the threshold for emotional classification, the text sentiment is defined as negative; and when f(w) is equal to the threshold for emotional classification, the text sentiment is defined as neutral.

[0082] Define the corresponding sentiment score for each text sentiment tendency based on f(w);

[0083] The sentiment scores in all the user's texts are summed and aggregated, and the normalization function is applied to obtain the user's overall sentiment tendency score, which is expressed as,

[0084]

[0085] Where S(u) represents the overall sentiment score of user u, tanh represents the hyperbolic tangent function, and T u represents the set of all texts of user u, where u represents the user, c(w) represents the sentiment tendency of the text, and s c(w)represents the sentiment score corresponding to c(w), β represents the adjustment parameter of the contribution of each text sentiment score. It should be noted that the adjustment parameter β of the contribution of each text sentiment score is determined by setting the initial value of β based on the knowledge and experience of domain experts, and then fine-tuning it according to the actual application effect.

[0086] It should be noted that the basis for designing the formula for calculating the user's overall sentiment tendency score is to aggregate the sentiment scores of all the user's texts and apply a normalization function to comprehensively reflect the user's overall sentiment tendency, and to adjust the contribution of each text sentiment score by adjusting the parameter β.

[0087] Based on each user's historical emotional tendency, set the overall emotional category division threshold;

[0088] When S(u) is greater than the overall sentiment category division threshold, the user's sentiment tendency is defined as positive;

[0089] When S(u) is less than the overall emotion classification threshold, the user's emotion tendency is defined as negative;

[0090] When S(u) is equal to the overall emotion classification threshold, the user's emotion tendency is defined as neutral.

[0091] It should be noted that by analyzing users' social media activities, comments and feedback content, using NLP (natural language processing) technology to identify user emotional tendencies, and deeply understand user needs, providing a basis for subsequent personalized recommendations, the accuracy of user portraits is improved, and the targeted nature of the course recommendation system is enhanced.

[0092] S3. Provide personalized course recommendations based on the recognition results of users’ emotional tendencies.

[0093] In the interval where the user's overall sentiment tendency is positive, provide personalized courses to the user and calculate the recommendation score of each personalized course, which is expressed as,

[0094]

[0095] Among them, R c represents the personalized course recommendation score of course c, U represents the user set, and s c represents the sentiment tendency score of course c, which can be obtained by sentiment analysis of the comments and feedback of the course, λ represents the parameter that adjusts the steepness of the function, which is used to control the shape of the function, and μ represents the average sentiment tendency score of all courses, which is used to calculate the relevance score of each user to the course;

[0096] It should be noted that μ is the average of the sentiment scores of all courses, which can be directly obtained by calculating the average of the sentiment scores of all courses. First, the sentiment score of each course is calculated, and then the sentiment scores of all courses are averaged to obtain μ, which reflects the average level of the sentiment scores of the entire course set. As a baseline, it helps to measure whether the sentiment score of a single course is higher or lower than the average level. λ is usually a positive number used to adjust the steepness of the function. In the recommendation system, it is used to control the sensitivity of the recommendation scoring function. The value of λ can be determined based on experiments. The best value is usually found through cross-validation. For example, a series of different λ values ​​(such as 0.1, 1, 10, etc.) can be set, and then the recommendation effect under different λ values ​​can be observed through A / B testing or through training sets / validation sets, so as to select the optimal λ value. If you want the recommendation results to be more sensitive to the user's sentiment, you can choose a larger λ value; conversely, if you want the recommendation results to be smoother, you can choose a smaller λ value.

[0097] It should be noted that the formula for calculating each personalized course recommendation score is designed based on calculating the correlation between each user's sentiment tendency score and the course sentiment tendency score, and using the adjustment parameter λ to control the steepness of the function, thereby obtaining a personalized recommendation score for each course;

[0098] According to the calculated recommendation score of each personalized course, personalized courses are sorted in descending order, and the top k personalized courses are recommended to the user.

[0099] It should be noted that based on the results of user sentiment tendency identification, personalized course recommendation scores are calculated, and courses are recommended accordingly, providing highly personalized course selections to meet the specific needs of users, increase user satisfaction and stickiness, and improve the competitiveness of the platform.

[0100] S4. Evaluate the user's yoga skill level and use reinforcement learning algorithms to plan the best learning path for the user.

[0101] Design a questionnaire survey on the yoga class booking platform, including basic information: collect basic information such as age, gender, occupation, etc. of users; Yoga experience: ask users whether they have experience in yoga practice, and if so, understand the length, frequency and type of practice; Skill self-assessment: let users self-assess their skill level in yoga, which can be divided into three levels: beginner, intermediate and advanced; Health status: ask users about their health status, including whether they have any health problems or restrictions and goals: such as stress relief, body shaping, improving flexibility, etc., and fully understand users' needs through various methods (including online surveys, face-to-face interviews and telephone interviews, etc.);

[0102] Based on the questionnaire data filled out by users, skill level evaluation indicators (such as asana mastery, breathing technique proficiency, balance ability, etc.) are designed, and evaluation standards are formulated in combination with various aspects of yoga skills (for example, for asana mastery, a scoring system from 1 to 5 can be set, where 1 means only the most basic asana can be completed, and 5 means the ability to complete advanced asanas) and the user's skill level score is calculated, expressed as,

[0103]

[0104] Among them, L u represents the skill level score of user u, n represents the number of evaluation indicators, x i represents the user's score on the i-th evaluation indicator, σ i The standard deviation of the i-th evaluation indicator is used to reflect the dispersion of the score of the indicator, where i represents the number of evaluation indicators;

[0105] It should be noted that σ i It represents the standard deviation of the ith evaluation indicator, which is used to reflect the degree of dispersion of the score of the indicator. The scores of all users on the ith evaluation indicator are collected, and the average score of all users on the ith evaluation indicator is calculated. The variance of the ith evaluation indicator is calculated first, and then the standard deviation of the ith evaluation indicator is calculated, which is expressed as,

[0106]

[0107] Among them, μ i ′ represents the average score of the i-th evaluation indicator;

[0108] Set skill level indicators based on professional knowledge of yoga skills;

[0109] When L u When it is greater than the skill level indicator, the user's yoga skill level is defined as advanced;

[0110] When L u When the skill level is less than or equal to the skill level indicator, the user's yoga skill level is defined as elementary;

[0111] Furthermore, based on the expertise in yoga skills and the knowledge of the users’ yoga skills during the questionnaire survey, we assume that L u The value range of is [0, 1], and the skill level can be divided into more specific ones:

[0112] When 0≤L u When <0.33, it means that the user is at the beginner level in terms of yoga skills. This range means that the user has just started to learn yoga and has only a preliminary understanding of the basic knowledge and asanas of yoga.

[0113] When 0.33≤L u When <0.66, it means that the user is at the beginner level. This range means that the user has mastered some basic knowledge of yoga and can complete some simple asanas.

[0114] When 0.66≤L u When <0.85, it means that the user is at an intermediate level. This range means that the user has good yoga skills and can complete most asanas.

[0115] When 0.85≤L u When <0.95, it means that the user is at an advanced level. This range means that the user is already quite proficient in yoga skills and can complete more difficult asanas.

[0116] When 0.95≤L u When ≤1, it means that the user has reached the expert level. This range means that the user has reached a very high level of yoga skills and can complete all asanas, including advanced asanas.

[0117] Define the state space and action space of the learning path. The state space includes factors such as the user's skill level, health status (whether the user has any health problems or restrictions, such as heart disease, high blood pressure, etc.), practice frequency (such as once a week, 2-3 times, 4-5 times, every day, etc.), practice time length (such as 30 minutes, 45 minutes, 60 minutes, etc.), practice preferences (such as Yin yoga, flow yoga, Hatha yoga, etc.). According to the user's yoga skill level, it can be divided into five levels: beginner, elementary, intermediate, advanced and expert. According to the user's health status, it can be divided into four levels: healthy, slightly restricted, moderately restricted, and severely restricted. According to the frequency of practice, it can be divided into three levels: low frequency, medium frequency, and high frequency. According to the length of practice time, it can be divided into short time, There are three levels: medium and long. Yoga can be divided into different types according to user preferences. The action space includes yoga courses of different types (such as Yin Yoga, Flow Yoga, Hatha Yoga, etc.) and difficulty levels (the difficulty level of each yoga course can be divided into three levels: elementary, intermediate and advanced). The reward function can be designed based on factors such as course completion, skill improvement or health improvement. For course completion, the reward score is adjusted according to the difficulty level of the course. The more difficult the course, the higher the reward score. For skill improvement, when the user's skill level score improves, the reward score increases, otherwise it decreases. For example, for health improvement, if the user's heart rate slows down or blood pressure decreases, it can be regarded as an improvement in health, which aims to encourage users to complete a higher quality learning path.

[0118] The DQN (Deep Q Network) algorithm is used to plan the best learning path. According to the user's state and the action taken, the state transition probability is defined, and the ε-greedy strategy is used to balance exploration and utilization, which is expressed as,

[0119]

[0120] Among them, Q(s,a) represents the expected cumulative reward for taking action a in state s, a represents the learning rate, which is used to control the influence of new information, r represents the immediate reward, which is given according to the completion of the course by the user, γ represents the discount factor, which is used to balance short-term and long-term rewards, and s ′ Represents the new state after taking action a, a ′ Represents the new state s ′ The next action, i.e. the course for the next learning stage, θ - It represents the parameters of the target network, which is used to stabilize the learning process and prevent instability during the learning process. s represents the state.

[0121] When selecting an action, an action is randomly selected with probability ∈ for exploration, and the action with the largest expected cumulative reward in the current state is selected with probability 1-∈ to plan the best learning path for the user.

[0122] It should be noted that by evaluating the user's yoga skill level, using the DQN algorithm to plan the best learning path, and providing users with customized learning plans, it promotes skill improvement, improves learning efficiency and personal progress, and enhances user participation.

[0123] S5. Users book courses based on the recommended course list and complete the payment process through blockchain technology.

[0124] Based on the user's personalized course recommendation and learning path planning, the most suitable yoga course list is recommended to the user. The user chooses the course he wants to book according to his needs and schedule. After the user confirms the booking information, the yoga booking platform generates a payment request. The user completes the payment verification through the blockchain wallet. After the payment is completed, the transaction record is added to the blockchain to ensure that the transaction cannot be tampered with;

[0125] In the payment process, Solidity (a high-level programming language specially designed for Ethereum smart contract development and used to write secure and reliable smart contracts) is used to write smart contracts to protect the smooth completion of the transaction process and ensure the security and transparency of payment.

[0126] It should be noted that users book courses and complete payments through blockchain technology, ensuring the security and transparency of the payment process, reducing the risk of fraud, and improving payment efficiency and trust.

[0127] S6. Submit feedback after the course to improve personalized course recommendations and learning path planning.

[0128] After arriving at the yoga class, the user signs in on the yoga class booking platform’s app. After the class, the user submits feedback on the yoga class booking platform (including the difficulty of the class, the instructor’s performance, personal feelings about the class, suggestions or improvement suggestions, and the course content).

[0129] Perform missing value processing and outlier detection on feedback data submitted by users, remove invalid or duplicate data, use the BERT model and the user's overall sentiment tendency score formula to perform sentiment analysis on open feedback, quantify user satisfaction, calculate the user's feedback score according to the feedback scoring formula, and adjust the weight of the feedback dimension according to the importance of the feedback content;

[0130] Add the new feedback data to the model's training dataset and retrain the deep reinforcement learning model using the updated dataset;

[0131] Based on the new model output, update the recommendation algorithm, adjust the recommendation strategy according to user feedback and changes in skill level, and update the learning path based on user feedback and new recommendation strategies.

[0132] It should be noted that user feedback is collected to improve personalized course recommendations and learning path planning, continuously optimize recommendation algorithms, improve service quality, enhance the system's self-learning and adaptability, and continuously improve user experience.

[0133] In summary, the present invention verifies the user identity through the zero-knowledge proof technology based on blockchain, which not only protects the user's privacy but also improves the trust of the platform. The user's social media activities, comments and feedback content are analyzed using natural language processing technology to identify their emotional tendencies, and then highly personalized course recommendations are provided, which greatly improves user satisfaction and stickiness. By evaluating the user's yoga skill level and using the reinforcement learning algorithm to plan the best learning path, the user's skill improvement and personal progress are effectively promoted. The use of blockchain technology to complete the payment process not only ensures the security and transparency of the transaction, but also reduces the risk of fraud and improves payment efficiency. By collecting user feedback and continuously improving personalized course recommendations and learning path planning, the system's self-learning and adaptability are enhanced, and the service quality is further improved. The yoga class appointment platform has become a safe, reliable, highly personalized and constantly improving service platform.

[0134] Example 2

[0135] Referring to Table 1, which is the second embodiment of the present invention, in order to further verify the advancement of the present invention, experimental simulation data of the intelligent shared yoga class booking platform system are provided.

[0136] This embodiment aims to evaluate the effects of blockchain-based identity authentication technology, sentiment recognition, personalized course recommendation, and learning path planning in the yoga class booking platform through a series of experiments. The experimental subjects are active users on the yoga class booking platform.

[0137] 200 users from different backgrounds who had never used a yoga class booking platform before were recruited.

[0138] The implementation steps include users registering an account on the yoga booking platform and verifying their identity through blockchain-based zero-knowledge proof technology. Analyze the content of users' social media activities, comments and feedback, and use natural language processing technology to identify users' emotional tendencies. Provide personalized course recommendations based on the identification results of users' emotional tendencies. Evaluate users' yoga skill levels and use reinforcement learning algorithms to plan the best learning path for users. Users book courses based on the recommended course list and complete the payment process through blockchain technology. Submit feedback after the course to improve personalized course recommendations and learning path planning.

[0139] The registration information, sentiment tendency score, skill level score, course reservation record, payment record and feedback score of each user were collected; statistical methods were used to compare the differences in engagement, satisfaction, etc. between users who used the above technologies and those who did not.

[0140] The details are shown in Table 1:

[0141] Table 1 Experimental record table

[0142]

[0143] In this example, various data of 200 users were collected through the above experimental design and analyzed in detail. The following are the key findings:

[0144] All 200 users successfully passed the identity authentication based on blockchain zero-knowledge proof technology, with a verification rate of 100%; users with high sentiment tendency scores (>0.80) are more inclined to book more courses, with an average number of courses booked being 3.5; in contrast, users with lower sentiment tendency scores (<0.70) book an average of 2.3 courses; users with higher skill level scores (>0.80) receive an average of 5 personalized recommended courses, while users with lower skill level scores (<0.60) receive an average of 3 personalized recommended courses; the average feedback score of users with a high number of payments (>3 times) is 4.6 points, and the average feedback score of users with a low number of payments (<=1 time) is 3.9 points.

[0145] In summary, this example shows that the yoga class booking platform significantly improves user engagement, satisfaction, and the number of course bookings by adopting blockchain-based identity authentication technology, sentiment recognition, personalized course recommendations, and learning path planning. In particular, users with high sentiment scores are more willing to book courses, and users with high skill level scores receive more personalized recommended courses. In addition, there is a clear positive correlation between the number of payments and feedback scores, indicating that the convenience of the payment process directly affects user satisfaction.

[0146] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. An intelligent shared yoga class booking platform system, characterized by: include, The user registration and identity authentication module is responsible for users to register accounts on the yoga booking platform and verify their identities through blockchain-based zero-knowledge proof technology; Sentiment analysis module, which is responsible for analyzing users’ social media activities, comments, and feedback, and using natural language processing technology to identify users’ sentiment tendencies; The personalized course recommendation module is responsible for providing personalized course recommendations based on the recognition results of the user's emotional tendencies; The learning path planning module is responsible for evaluating the user's yoga skill level and using reinforcement learning algorithms to plan the best learning path for the user; The course reservation and payment module is responsible for users to reserve courses based on the recommended course list and complete the payment process through blockchain technology; The feedback collection and optimization module is responsible for submitting feedback after the course ends to improve personalized course recommendations and learning path planning.

2. The intelligent shared yoga class booking platform system according to claim 1, characterized in that: Users register an account on the yoga class booking platform and verify their identity through blockchain-based zero-knowledge proof technology, including the following steps: The user visits the yoga class booking platform, clicks the registration button to enter the registration page, fills in the registration information, and completes the registration; After the user completes the registration, he / she enters the identity verification page. The platform uses blockchain-based zero-knowledge proof technology to verify the user's identity information. After the verification is passed, the platform generates a unique blockchain identifier for the user, which is associated with the user's account. If the verification fails, modify the incorrect user information and re-verify until the verification passes.

3. The intelligent shared yoga class booking platform system according to claim 2, characterized in that: Analyze the content of users’ social media activities, comments and feedback, including the following steps: Collect text data of users’ social media activities, comments, and feedback, split the text data into individual words, and remove non-text information; The TF-IDF value of each word is calculated using TF-IDF technology to obtain text features, and the feature representation is further enhanced using word embedding technology.

4. The intelligent shared yoga class booking platform system according to claim 3, characterized in that: Using natural language processing technology to calculate the user's overall sentiment tendency score includes the following steps: Use the BERT model to perform sentiment analysis on each enhanced text feature, expressed as, f(w) = N(w; θ); Where N represents the BERT model, w represents the enhanced text features, θ represents the parameters of the model, and f(w) represents the sentiment category score; Based on each user's historical emotional tendency, set the emotional category classification threshold; When f(w) is greater than the threshold for emotional classification, the text sentiment is defined as positive; when f(w) is less than the threshold for emotional classification, the text sentiment is defined as negative; and when f(w) is equal to the threshold for emotional classification, the text sentiment is defined as neutral. Define the corresponding sentiment score for each text sentiment tendency based on f(w); The sentiment scores in all the user's texts are summed and aggregated, and the normalization function is applied to obtain the user's overall sentiment tendency score, which is expressed as, Where S(u) represents the overall sentiment score of user u, tanh represents the hyperbolic tangent function, and T u represents the set of all texts of user u, where u represents the user, c(w) represents the sentiment tendency of the text, and s c(w) represents the sentiment score corresponding to c(w), and β represents the adjustment parameter of the contribution of each text sentiment score.

5. The intelligent shared yoga class booking platform system according to claim 4, characterized in that: Based on the user's overall sentiment tendency score, the user's sentiment tendency is identified, including the following steps: Based on each user's historical emotional tendency, set the overall emotional category division threshold; When S(u) is greater than the overall sentiment category division threshold, the user's sentiment tendency is defined as positive; When S(u) is less than the overall emotion classification threshold, the user's emotion tendency is defined as negative; When S(u) is equal to the overall emotion classification threshold, the user's emotion tendency is defined as neutral.

6. The intelligent shared yoga class booking platform system according to claim 5, characterized in that: Based on the recognition results of the user's emotional tendency, personalized course recommendations are provided, including the following steps: In the interval where the user's overall sentiment tendency is positive, provide personalized courses to the user and calculate the recommendation score of each personalized course, which is expressed as, Among them, R c represents the personalized course recommendation score of course c, U represents the user set, and s c represents the sentiment tendency score of course c, λ represents the parameter that adjusts the steepness of the function, and μ represents the average sentiment tendency score of all courses; According to the calculated recommendation score of each personalized course, personalized courses are sorted in descending order, and the top k personalized courses are recommended to the user.

7. The intelligent shared yoga class booking platform system according to claim 6, characterized in that: Use questionnaires to understand users and assess their yoga skill levels, including the following steps: Design a questionnaire survey on the yoga class booking platform to fully understand user needs through various methods; Based on the questionnaire data filled out by users, we design a skill level evaluation index, formulate an evaluation standard based on various aspects of yoga skills, and calculate the user's skill level score, which is expressed as, Among them, L u represents the skill level score of user u, n represents the number of evaluation indicators, x i represents the user's score on the i-th evaluation indicator, σ i The standard deviation of the i-th evaluation indicator, where i represents the number of evaluation indicators; Set skill level indicators based on professional knowledge of yoga skills; When L u When it is greater than the skill level indicator, the user's yoga skill level is defined as advanced; When L u When the skill level is less than or equal to the skill level indicator, the user's yoga skill level is defined as elementary.

8. The intelligent shared yoga class booking platform system according to claim 7, characterized in that: Based on the user's yoga skill level, the reinforcement learning algorithm is used to plan the best learning path for the user, including the following steps: Define the state space and action space of the learning path, and design a reward function to encourage users to complete a higher quality learning path; Use the DQN algorithm to plan the best learning path, define the state transition probability according to the user's state and the action taken, and use the ε-greedy strategy to balance exploration and utilization, expressed as, Among them, Q(s,a) represents the expected cumulative reward of taking action a in state s, a represents the learning rate, r represents the immediate reward, γ represents the discount factor, s ′ Represents the new state after taking action a, a ′ Represents the new state s ′ The next action, θ - represents the parameters of the target network, and s represents the state; When selecting an action, an action is randomly selected with probability ∈ for exploration, and the action with the largest expected cumulative reward in the current state is selected with probability 1-∈ to plan the best learning path for the user.

9. The intelligent shared yoga class booking platform system according to claim 8, characterized in that: Users book courses based on the recommended course list and complete the payment process through blockchain technology, including the following steps: Based on the user's personalized course recommendation and learning path planning, the most suitable yoga course list is recommended to the user. The user chooses the course he wants to book according to his needs and schedule. After the user confirms the booking information, the yoga booking platform generates a payment request. The user completes the payment verification through the blockchain wallet. After the payment is completed, the transaction record is added to the blockchain; Use Solidity to write smart contracts in the payment process to ensure the smooth completion of the transaction.

10. The intelligent shared yoga class booking platform system according to claim 9, characterized in that: Submit feedback after the course to improve personalized course recommendations and learning path planning, including the following steps: After arriving at the yoga studio, users sign in on the yoga class booking platform’s app. After the class, users submit feedback on the yoga class booking platform. Clean the feedback data submitted by users, remove invalid or duplicate data, use natural language processing technology to perform sentiment analysis on open feedback, quantify user satisfaction, calculate user feedback scores based on the feedback scoring formula, and adjust the weights of feedback dimensions based on the importance of the feedback content; Add the new feedback data to the model's training dataset and retrain the deep reinforcement learning model using the updated dataset; Based on the new model output, update the recommendation algorithm, adjust the recommendation strategy according to user feedback and changes in skill level, and update the learning path based on user feedback and new recommendation strategies.

Images