A decentralized eloquence skill matching method, device and medium

Abstract

The present invention discloses a decentralized eloquence skill matching method, device and medium, the method comprising: obtaining speech data of several users; comprehensively evaluating the speech data to obtain skill evaluation results; training several users according to eloquence courses to obtain training data; in a decentralized autonomous organization, dividing the eloquence course into several assets, and matching several assets with several users according to the skill evaluation results and the training data to obtain matching results. The present invention proposes a decentralized eloquence skill matching method, device and medium, by comprehensively evaluating the speech data and training several users with the eloquence course, the speech strengths and weaknesses of different users and the degree of course adaptation can be fully understood, and then on the basis of the decentralized autonomous organization, which is a community participation decision-making, eloquence resources are matched with users, which can solve the problem of difficulty in combining community participation decision-making to achieve effective matching of eloquence resources with users.

Description

Technical Field

[0001] The present invention relates to the field of speech training, and in particular to a decentralized eloquence skill matching method, device and medium. Background Art

[0002] In the mechanism of matching eloquence resources with users, community members include users and resource providers. Their active participation can provide valuable feedback and suggestions for the matching process, which helps to understand user needs and market changes more accurately, thereby improving the efficiency and accuracy of eloquence resource matching; and the direct participation of community members means that matching decisions are more in line with their expectations and needs, which can not only improve user satisfaction, but also enhance their trust and sense of belonging to the platform. The traditional matching method is to invite residents to participate in the evaluation and supervision of eloquence resource and user matching projects by setting up specific evaluation and supervision teams to ensure that the matching results meet the needs and expectations of the community.

[0003] However, although the importance of community participation in decision-making is widely recognized, in actual operations, due to imperfect participation mechanisms, resource allocation is relatively subjective and prone to deviations. This unbalanced resource allocation method will lead to unreasonable distribution of eloquence resources within the community, resulting in some users who are capable but in need of resources may not be able to obtain sufficient and appropriate support. Summary of the invention

[0004] The present invention provides a decentralized eloquence skill matching method, device and medium to solve the problem that it is difficult to combine community participation in decision-making and achieve effective matching of eloquence resources and users.

[0005] Get speech data of several users;

[0006] Performing a comprehensive evaluation on the speech data to obtain a skill evaluation result;

[0007] Training the plurality of users according to the eloquence course to obtain training data;

[0008] In a decentralized autonomous organization, the eloquence course is divided into a plurality of assets, and the plurality of assets are matched with the plurality of users according to the skill evaluation result and the training data to obtain a matching result;

[0009] The weight set of the centralized autonomous organization is obtained by adjusting the initial asset voting weights of the users in the centralized autonomous organization according to the skill evaluation results.

[0010] The present invention can preliminarily understand the user's speech ability and know the user's strengths and weaknesses in speech by comprehensively evaluating the user's speech data; according to the systematic eloquence course, the user can be trained in a targeted manner for his or her oral expression ability, logical thinking ability, psychological quality, etc., further reflecting the individual differences and advantages and disadvantages of the user in eloquence training, and the training data directly records the user's performance in the eloquence course, reflecting the user's absorption degree of the course content, adaptability to the teaching method, and completion of the training task, so by observing these data, it can be known whether the course is suitable for the user's actual level and needs, revealing the preferences and adaptability of different users to different parts of the course. Since the weight set is obtained by adjusting the asset voting weights of several users in the community autonomous organization according to the skill evaluation results, the weight set can ensure that more experienced and skilled users have greater influence in key decisions on asset importance, so that user opinions can participate in community decision-making as an effective factor. Therefore, in the decentralized autonomous organization, matching several assets with several users according to the skill evaluation results and training data can fully combine community participation in decision-making, and match the corresponding assets that are currently most suitable for the user by considering the differences in user speech abilities.

[0011] Compared with the prior art, the present invention can fully understand the speech strengths and weaknesses of different users and the degree of course adaptability by comprehensively evaluating the speech data and using eloquence courses to train several users, and then objectively match eloquence resources with users on the basis of community participation decision-making in decentralized autonomous organizations. Therefore, it can solve the problem of difficulty in combining community participation decision-making to achieve effective matching of eloquence resources with users.

[0012] As a preferred solution, a comprehensive evaluation is performed on the speech data to obtain a skill evaluation result, specifically:

[0013] According to the emotion intensity and tone intensity of the speech data at different times, respectively calculating the expressiveness index and the tone variation degree;

[0014] Calculating language richness and knowledge depth respectively based on the vocabulary usage frequency and knowledge point usage accuracy of the speech data;

[0015] The skill evaluation result is composed of the expressiveness index, the pitch variation, the language richness and the knowledge depth.

[0016] In this preferred solution, the expressiveness index can measure the clarity and effectiveness of the information conveyed by the user; the pitch variation measures the variation and flexibility of the user's voice. A higher pitch variation may indicate that the user can attract the audience's attention and enhance the effect of expression through the ups and downs of the voice; the language richness measures the diversity and complexity of the user's language through the frequency of vocabulary use; the knowledge depth can evaluate the user's knowledge mastery in a certain field or topic, reflecting the user's understanding and professional quality in the field; therefore, the comprehensive evaluation results of these four aspects can form a comprehensive evaluation of the user's eloquence or presentation skills, reflecting the user's comprehensive presentation ability.

[0017] As a preferred solution, the eloquence course is specifically:

[0018] According to the adaptability of the plurality of users to the preset courses, an optimization problem regarding course adjustment is established;

[0019] By simulating the skill changes of the plurality of users, a change value of eloquence skill is obtained;

[0020] The speech data is mapped to an evaluation space of user performance in an algebraic structure, and the eloquence course is established by combining the optimization problem and the eloquence skill change value; wherein the algebraic structure is a mathematical structure configured by the preset course.

[0021] This preferred solution can determine the goal of course adjustment by analyzing the causes of inadaptability, that is, how to reduce or eliminate these inadaptabilities by changing course content, teaching methods or progress. By simulating the changes in users' skills, the impact of different teaching methods, course content or progress on users' eloquence skills can be predicted or evaluated, which helps to understand which teaching method or content adjustment is more likely to promote user skill improvement. Mapping eloquence data to the evaluation space of user performance in an algebraic structure can transform qualitative or complex data into a form that can be quantified, analyzed and compared, thereby providing more accurate guidance for course adjustment; this way of establishing eloquence courses can ensure that course content and teaching methods can meet user needs to the greatest extent possible.

[0022] As a preferred solution, according to the adaptability of the plurality of users to the preset courses, an optimization problem regarding course adjustment is established, specifically:

[0023] Define the desired state to be achieved by the end of the pre-set course;

[0024] Establishing a terminal cost function by quantifying the gap between the state at the end of the preset course and the ideal state;

[0025] Establishing a loss function according to the inadaptability losses of the plurality of users to the course configuration in the preset course;

[0026] According to the terminal cost function, the course configuration and the loss function, the optimization problem regarding course adjustment is established.

[0027] This preferred solution can clearly see the degree of deviation between the current course configuration and the ideal state by quantifying the gap between the state at the end of the course and the ideal state. The terminal cost function is a mathematical tool for evaluating the gap between the state at the end of the course and the ideal state. Through this function, the effect of the current course configuration can be more accurately described, providing quantitative guidance for course adjustment. A loss function is established based on the inadaptability loss of several users to the course configuration, which can provide in-depth understanding of users' feedback and opinions on the current course configuration. Combining the course configuration, loss function and terminal cost function, an optimization problem for course adjustment is constructed. This optimization problem is actually a mathematical model. By solving the optimization problem, a set of optimal course configuration solutions can be obtained. These solutions can guide course designers to improve the current course configuration to improve the teaching effect of the course and user satisfaction.

[0028] As a preferred solution, by simulating the skill changes of the several users, the eloquence skill change value is obtained, specifically:

[0029] Simulating the skill changes of the plurality of users according to a preset partial differential equation to obtain the eloquence skill change value;

[0030] Wherein, the partial differential equation is:

[0031]

[0032] Among them, s(x,t) represents the skill level at position x and time t, D is the diffusion coefficient, λ is the attenuation coefficient, and f(x,t) is the parameter input by external training.

[0033] This preferred solution can accurately simulate the changes in users' eloquence skills over time or other factors through preset partial differential equations, providing strong data support for subsequent course adjustments.

[0034] As a preferred solution, the plurality of users are trained according to the eloquence course to obtain training data, specifically:

[0035] Using the eloquence course to train the plurality of users to obtain initial training data;

[0036] Measuring the speech skill change of the initial training data to obtain skill data;

[0037] Measuring the association between different speech skills in the initial training data by complex variable mapping and continuous mapping to obtain association data;

[0038] Based on the initial training data, simulating the spatial distribution and change trend of speech skills to obtain change data;

[0039] Calculate rewards based on the course activity and learning progress of the several users in the initial training data;

[0040] The training data is composed of the evaluation data of the course, the skill data, the association data, the change data and the reward.

[0041] This preferred solution obtains skill data by measuring the changes in speech skills of the initial training data. These data reflect the rise and fall of the user's speech skills; by measuring the relationship between different speech skills to obtain correlation data, the internal connection and mutual influence between different skills are revealed, which helps to understand the overall structure and development laws of speech skills. Based on the initial training data, the spatial distribution and change trend of speech skills are simulated to obtain change data. These data show the dynamic changes of speech skills at different times and between different users, which helps to predict and plan future skill training strategies. Rewards are calculated based on the user's active participation in the course and learning progress, which can effectively motivate users to participate in learning more actively and improve learning motivation and results. Combining these data to form training data can reflect the user's multi-faceted information about eloquence skill training and provide a rich data source for subsequent asset matching.

[0042] As a preferred solution, after calculating the reward according to the course activity and learning progress of the plurality of users in the initial training data, the method further includes:

[0043] Evaluating eloquence performance based on the initial training data, and using the evaluation results to adjust the incentive mechanism of the centralized autonomous organization;

[0044] According to the topological structure of the initial training data, controlling the reward allocation algorithm of the centralized autonomous organization to obtain the internal structure of the initial training data;

[0045] The integrity and consistency of the data in the incentive mechanism are ensured by a preset abstract algebraic structure.

[0046] In this preferred solution, the incentive mechanism is an important driving force for users to actively participate and improve their skills in centralized autonomous organizations. Adjusting the incentive mechanism of centralized autonomous organizations based on the eloquence performance evaluation results of the initial training data can ensure the effective allocation of resources and promote the improvement of overall learning effects. The topological structure of the initial training data reveals information such as skill connections and interaction patterns between users. Controlling the reward allocation algorithm to obtain this internal structure can ensure that rewards are not only based on individual performance, but also take into account factors such as individual contributions and influence in the community, which helps promote the healthy development of the community and enhance cooperation and interaction between users.

[0047] As a preferred solution, the evaluation data of the course is specifically:

[0048] According to the initial training data, the comprehensive quality of eloquence is calculated by means of multiple regression analysis to obtain a comprehensive value;

[0049] According to the time series data in the initial training data, a dynamic change value of the course evaluation is calculated by a differential equation;

[0050] Measuring the effects and relationships between the evaluation indicators in the initial training data by using partial derivatives of higher-order functions to obtain relationship data;

[0051] The evaluation data is composed of the comprehensive value, the dynamic change value and the relationship data.

[0052] In this preferred solution, the time series data in the initial training data contains the course evaluation information of users at different time points. The dynamic change value of the course evaluation is calculated by differential equations, which reflects the trend of course evaluation over time, which is of great significance for understanding the long-term effect of the course and predicting the future development trend of the course. There may be complex interactions and relationships between the evaluation indicators in the initial training data. Obtaining relational data through partial derivatives of higher-order functions helps to deeply understand the internal connection between the evaluation indicators and provide direction for subsequent improvement and optimization.

[0053] The present invention also provides a decentralized eloquence skill matching device, comprising a data module, an evaluation module, a training module and a matching module;

[0054] Wherein, the data module is used to obtain speech data of several users;

[0055] The evaluation module is used to comprehensively evaluate the speech data to obtain a skill evaluation result;

[0056] The training module is used to train the plurality of users according to the eloquence course to obtain training data;

[0057] The matching module is used to divide the eloquence course into a plurality of assets in the decentralized autonomous organization, and match the plurality of assets with the plurality of users according to the skill evaluation result and the training data to obtain a matching result;

[0058] The weight set of the centralized autonomous organization is obtained by adjusting the initial asset voting weights of the users in the centralized autonomous organization according to the skill evaluation results.

[0059] As a preferred solution, the evaluation module includes an emotion unit, a knowledge unit and a skill unit;

[0060] The emotion unit is used to calculate the expression index and the pitch change degree according to the emotion intensity and pitch intensity of the speech data at different times;

[0061] The knowledge unit is used to calculate the language richness and knowledge depth respectively according to the vocabulary usage frequency and knowledge point usage accuracy of the speech data;

[0062] The skill unit is used to form the skill evaluation result by the expressiveness index, the pitch variation, the language richness and the knowledge depth.

[0063] As a preferred solution, the eloquence course is specifically:

[0064] According to the adaptability of the plurality of users to the preset courses, an optimization problem regarding course adjustment is established;

[0065] By simulating the skill changes of the plurality of users, a change value of eloquence skill is obtained;

[0066] The speech data is mapped to an evaluation space of user performance in an algebraic structure, and the eloquence course is established by combining the optimization problem and the eloquence skill change value; wherein the algebraic structure is a mathematical structure configured by the preset course.

[0067] As a preferred solution, according to the adaptability of the plurality of users to the preset courses, an optimization problem regarding course adjustment is established, specifically:

[0068] Define the desired state to be achieved by the end of the pre-set course;

[0069] Establishing a terminal cost function by quantifying the gap between the state at the end of the preset course and the ideal state;

[0070] Establishing a loss function according to the inadaptability losses of the plurality of users to the course configuration in the preset course;

[0071] According to the terminal cost function, the course configuration and the loss function, the optimization problem regarding course adjustment is established.

[0072] As a preferred solution, by simulating the skill changes of the several users, the eloquence skill change value is obtained, specifically:

[0073] Simulating the skill changes of the plurality of users according to a preset partial differential equation to obtain the eloquence skill change value;

[0074] Wherein, the partial differential equation is:

[0075]

[0076] Among them, s(x,t) represents the skill level at position x and time t, D is the diffusion coefficient, λ is the attenuation coefficient, and f(x,t) is the parameter input by external training.

[0077] As a preferred solution, the training module includes a training unit, a change unit, a mapping unit, a distribution unit, a reward unit and a comprehensive unit;

[0078] The training unit is used to train the plurality of users using the eloquence course to obtain initial training data;

[0079] The change unit is used to measure the speech skill change of the initial training data to obtain skill data;

[0080] The mapping unit is used to measure the association between different speech skills in the initial training data through complex variable mapping and continuous mapping to obtain association data;

[0081] The distribution unit is used to simulate the spatial distribution and change trend of speech skills based on the initial training data to obtain change data;

[0082] The reward unit is used to calculate a reward according to the course activity and learning progress of the plurality of users in the initial training data;

[0083] The integration unit is used to form the training data from the course evaluation data, the skill data, the association data, the change data and the reward.

[0084] As a preferred solution, after the reward unit, it also includes an incentive unit, a structural unit and a consistency unit;

[0085] The incentive unit is used to evaluate the eloquence performance according to the initial training data, and use the evaluation results to adjust the incentive mechanism of the centralized autonomous organization;

[0086] The structural unit is used to control the reward allocation algorithm of the centralized autonomous organization to obtain the internal structure of the initial training data according to the topological structure of the initial training data;

[0087] The consistency unit is used to ensure the integrity and consistency of the data in the incentive mechanism through a preset abstract algebraic structure.

[0088] As a preferred solution, the evaluation data of the course is specifically:

[0089] According to the initial training data, the comprehensive quality of eloquence is calculated by means of multiple regression analysis to obtain a comprehensive value;

[0090] According to the time series data in the initial training data, a dynamic change value of the course evaluation is calculated by a differential equation;

[0091] Measuring the effects and relationships between the evaluation indicators in the initial training data by using partial derivatives of higher-order functions to obtain relationship data;

[0092] The evaluation data is composed of the comprehensive value, the dynamic change value and the relationship data.

[0093] The present application also provides a storage medium, on which a computer program is stored. The computer program is called and executed by a computer to implement the decentralized eloquence skill matching method as described above. BRIEF DESCRIPTION OF THE DRAWINGS

[0094] Figure 1 It is a flowchart of a decentralized eloquence skill matching method provided in an embodiment of the present application;

[0095] Figure 2 It is a data flow diagram provided by an embodiment of the present application;

[0096] Figure 3 It is a structural diagram of a decentralized eloquence skill matching device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0097] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0098] In the description of the present application, unless otherwise specified, “several” means two or more.

[0099] A decentralized eloquence skill matching method provided in the embodiment of the present application is mainly used in situations where it is necessary to improve the mechanism of community participation in decision-making, and objectively match resources related to eloquence courses according to the user's own eloquence level and needs.

[0100] Embodiment 1:

[0101] See also Figure 1 The embodiment of the present application provides a decentralized eloquence skill matching method, including S1 to S4, and the specific implementation steps are as follows:

[0102] S1. Obtain speech data of several users.

[0103] Step S1 of the embodiment of the present application is specifically as follows:

[0104] Collect the speech data of several users, store the speech data in the skill archive database, and use blockchain technology to ensure the storage security and effective management of the data; among them, the speech data includes the basic information of several users (such as name, email, contact information, etc.) and eloquence-related information, and the skill archive database is used to specifically store the user's eloquence skill certification documents, such as certification, speech videos, etc.

[0105] In the preset DAO (Decentralized Autonomous Organization), a unique identity is created based on the basic information of the current user; the current user refers to the user who needs to obtain resource matching permissions to implement this solution.

[0106] S2. Comprehensively evaluate the speech data to obtain skill assessment results.

[0107] Step S2 of the embodiment of the present application is specifically as follows:

[0108] Fuzzy sets are established with expressiveness index, pitch variation, language richness and knowledge depth as category standards; and fuzzy sets allow variables to exist in a way of "belonging" or "not belonging" to a certain degree. The set can be divided into different fuzzy levels according to actual needs, such as "very weak", "weak", "medium", "strong" and "very strong", etc.; and a membership function is defined for the fuzzy set, which describes the degree to which the variable value belongs to the corresponding category of the set;

[0109] By mapping the speech data to the corresponding membership degree in the fuzzy set, the discrete or approximately discrete data in the speech data is converted into fuzzy values ​​to obtain a fuzzy value set;

[0110] The fuzzy value set is defuzzified by the maximum membership method to obtain a continuous data set;

[0111] With the help of the skill analysis engine, the expressiveness index EI(t) is calculated based on the emotion intensity and change rate of the continuous data set using the expressiveness index dynamic model EI(t). The skill analysis engine has built-in natural language processing and machine learning algorithms.

[0112] The pitch variation degree PT is calculated based on the pitch variation amplitude and flexibility of the continuous data set;

[0113] According to the vocabulary usage frequency and sentence structure of the continuous data set, the language richness LR is calculated;

[0114] The knowledge depth KD is calculated based on the coverage and usage accuracy of knowledge points in the continuous data set;

[0115] The skill evaluation results are composed of the expressiveness index EI, the pitch variation PT, the language richness LR and the knowledge depth KD, and the skill evaluation results are recorded in the user's blockchain profile;

[0116] Among them, the dynamic model of the expressiveness index is:

[0117]

[0118] The pitch variation is:

[0119]

[0120] The language richness is:

[0121]

[0122] The depth of knowledge is:

[0123]

[0124] Among them, E(t) represents the intensity of emotion at time t, α, β and γ are coefficients for adjusting the dynamic influence of emotion;

[0125] represents the intensity of the tone at time t and frequency f; represents the second-order partial derivative with respect to time, which is used to measure the acceleration of pitch change;

[0126] V i represents the usage frequency of the i-th category of words, λ i is the weight factor of the i-th category of vocabulary, and N is the total number of categories of vocabulary; represents the gradient operator used to find the vocabulary usage strategy that maximizes language richness;

[0127] K(t) represents the coverage of knowledge points at time t, A(t) represents the accuracy of the corresponding knowledge points at time t, and T is the total duration of the speech.

[0128] In this embodiment S2, the expressiveness index can measure the clarity and effectiveness of the information conveyed by the user; the pitch variation measures the variation and flexibility of the user's voice. A higher pitch variation may indicate that the user can attract the audience's attention through the fluctuation of the voice and enhance the effect of expression; the language richness measures the diversity and complexity of the user's language through the frequency of vocabulary use; the knowledge depth can evaluate the user's knowledge mastery in a certain field or topic, reflecting the user's understanding and professional quality in the field; therefore, the evaluation results of these four aspects can form a comprehensive evaluation of the user's eloquence or speech skills, reflecting the user's comprehensive speech ability;

[0129] In addition, this embodiment introduces calculus equations and partial derivative formulas of higher-order functions to describe and optimize continuous changes in the speech evaluation process, which can effectively increase the accuracy and flexibility of the model in processing continuous data streams and dynamic adjustments. By providing a transparent skill evaluation process, users can view and verify their evaluation history and results at any time.

[0130] S3. Train a number of users according to the eloquence course to obtain training data.

[0131] Step S3 of the embodiment of the present application includes S3.1 to S3.7; wherein S3.1 is a process of establishing an optimization problem for course adjustment, S3.2 is a process of continuously monitoring eloquence skills to obtain eloquence skill change values, S3.3 is a process of establishing an eloquence course based on the optimization problem and the eloquence skill change values, S3.4 is a process of calculating skill data, associated data and change data, S3.5 is a process of calculating rewards and making preliminary adjustments to the mechanism, S3.6 is a process of calculating evaluation data, and S3.7 is a process of generating training data based on evaluation data, skill data, associated data, change data and rewards, specifically:

[0132] S3.1. Define the ideal state you want to achieve by the end of the course;

[0133] By quantifying the gap between the state at the end of the preset course and the ideal state, the terminal cost function g(c(T)) is established;

[0134] The loss function L(c(t),t) is established based on the inadaptability loss of several users to the course configuration c(t) in the preset course;

[0135] According to the terminal cost function g(c(T)), the course configuration c(t) and the loss function L(c(t),t), an optimization problem F about course adjustment is established;

[0136] Among them, the optimization problem of course adjustment is:

[0137]

[0138] Among them, c(t) represents the course configuration at time t; L(c(t), t) is the loss function, which reflects the user's inadaptability to the current course configuration; g(c(T)) is the terminal cost function, which reflects the state at the end of the course; T is the total duration of the preset course.

[0139] This embodiment S3.1 can clearly see the degree of deviation between the current course configuration and the ideal state by quantifying the gap between the state at the end of the course and the ideal state. The terminal cost function is a mathematical tool for evaluating the gap between the state at the end of the course and the ideal state. Through this function, the effect of the current course configuration can be more accurately described, providing quantitative guidance for course adjustment. A loss function is established based on the inadaptability loss of several users to the course configuration, which can provide an in-depth understanding of the user's feedback and opinions on the current course configuration. An optimization problem for course adjustment is constructed by combining the course configuration, the loss function and the terminal cost function. This optimization problem is actually a mathematical model. By solving the optimization problem, a set of optimal course configuration solutions can be obtained. These solutions can guide course designers to improve the current course configuration to improve the teaching effect of the course and user satisfaction, and can meet the needs of real-time adjustment of course content.

[0140] S3.2, simulating the skill changes of several users according to the preset partial differential equation to obtain the eloquence skill change value;

[0141] The partial differential equation is:

[0142]

[0143] Among them, s(x,t) represents the skill level at position x and time t, D is the diffusion coefficient, λ is the attenuation coefficient, and f(x,t) is the parameter input by external training.

[0144] This embodiment S3.2 can accurately simulate the changes in the user's eloquence skills over time or other factors through the preset partial differential equation, provide strong data support for subsequent course adjustments, and meet the needs of real-time monitoring and updating of the user's eloquence skills.

[0145] S3.3, mapping the user performance in the speech data to the evaluation space of the user performance in the algebraic structure C, and establishing the eloquence course by combining the optimization problem and the eloquence skill change value; wherein the algebraic structure C is the mathematical structure configured by the preset course;

[0146] The algebraic structure is:

[0147]

[0148] The process of mapping user performance to evaluation space can be expressed as:

[0149] E:X→Y

[0150] Among them, C represents the algebraic structure of the entire course data, which is a set that contains the data of all course modules; C i It is the data representation of a single course module, such as a chapter, a lesson, or a specific skill training unit; i is the ordinal number of the course module, and n is the total number of course modules; The direct sum of the data structure is used to ensure the independence and overall consistency between data components. In this embodiment, the direct sum ensures that each course module C i While functioning as independent units, they together form a coherent structure for the entire Course C.

[0151] X is the set of user performances, Y is the space of user performance evaluation results; E represents a continuous mapping from X to Y, which is used to ensure the consistency and fairness of the evaluation.

[0152] By analyzing the causes of the inadaptability, this embodiment S3.3 can determine the goal of course adjustment, that is, how to reduce or eliminate these inadaptabilities by changing the course content, teaching methods or progress. By simulating the changes in users' skills, the impact of different teaching methods, course content or progress on users' eloquence skills can be predicted or evaluated, which helps to understand which teaching method or content adjustment is more likely to promote the user's skill improvement. Mapping eloquence data to the evaluation space of user performance in an algebraic structure can convert qualitative or complex data into a form that can be quantified, analyzed and compared, thereby providing more accurate guidance for course adjustments; this way of establishing eloquence courses can ensure that the course content and teaching methods can meet the needs of users to the greatest extent possible;

[0153] And, in abstract algebra, each component C i It has its own algebraic properties (such as the structure of a group, ring or field), so that data operations (such as adding, updating or deleting information) can be performed while maintaining structural and mathematical properties; this structured data processing method helps to ensure the correctness of operations and the integrity of data; the use of abstract algebraic structures can systematically manage course data, and the mathematical rules in it can ensure that all operations do not violate the defined data rules. For example, if course information is connected through a specific functional relationship (such as mapping or transformation), when the data is updated or modified, the related functions will also be updated accordingly, thus ensuring the consistency of the data.

[0154] S3.4, using the eloquence course to train a number of users to obtain initial training data; wherein the initial training data includes the feedback of the users on the course and their evaluation of the teacher;

[0155] Construct a Hilbert Space suitable for eloquence skill evaluation; according to the functional model Φ(f), the initial training data is converted into elements that can be represented in the Hilbert space by feature extraction and encoding; the changes in the initial training data in speech skills are measured by the inner product operation in the Hilbert space to obtain the skill data;

[0156] The complex correlation between different speech skills in the initial training data is measured by the complex variable mapping G(z), and the continuity and change between different speech skills in the initial training data are measured by the continuous mapping in topology to obtain the correlation data;

[0157] Based on the initial training data, according to the geometric surface model, the curve and surface theory in geometry is used to simulate the spatial distribution and change trend of speech skills to obtain change data;

[0158] Among them, the functional model is:

[0159] Φ(f)=∫ D (K(x,f(x))dx)

[0160] The complex variable mapping can be expressed as:

[0161]

[0162] A continuous map in topology can be represented as:

[0163] ψ:X→Y, continuous Opening ψ- 1 (V) is also an open set in X

[0164] The geometric surface model is:

[0165] S(t,θ)=(tcosθ,tsinθ,sinf(t))

[0166] Where f(x) represents a specific skill performance in the speech process, such as pitch or speaking speed; K(x,y) is a kernel function used to deal with nonlinear relationships in speech data; D is the time region of the speech duration;

[0167] f(ζ) represents a complex analytical function of eloquence performance, C is a closed path around the observation point z, and i is an imaginary unit;

[0168] X and Y represent the topological space of skill levels before and after speech, respectively;

[0169] t and θ represent the variation parameters of time and pitch, and f(t) represents the skill variation function related to time.

[0170] This embodiment S3.4 uses the complex analysis method to gain a deep understanding of the internal structure of speech skill changes, especially the data association when dealing with periodic and step changes, providing strong data support for subsequent asset matching. Since the initial training data includes feedback from several users on the course and evaluations of the teacher, it can also help other users select course assets.

[0171] S3.5. Based on the course activity and learning progress of several users in the initial training data, the incentive function model R(t) is used to calculate the impact of user behavior changes on eloquence improvement and obtain rewards;

[0172] Based on the initial training data, a functional space is constructed and the eloquence performance functional J[f] is used to evaluate the eloquence performance to obtain the evaluation results; the evaluation results and the optimization of the functional are used to adjust the incentive mechanism of the DAO;

[0173] The complex analysis model F(z) is used to process the complex patterns of user behavior in the initial training data, obtain the topological structure, and control the DAO's reward distribution algorithm to obtain and grasp the internal structure of the initial training data according to the topological structure;

[0174] The integrity and consistency of the data in the incentive mechanism are guaranteed through the preset abstract algebraic G structure; the transparency and fairness of the incentive mechanism are ensured by using smart contracts, making all reward records public and traceable;

[0175] Among them, the activation function model is:

[0176]

[0177] The eloquence performance functional is:

[0178]

[0179] The complex analysis model is:

[0180]

[0181] The abstract algebra is:

[0182]

[0183] Where A(t) is the user's activity at time t, P(t) is the learning progress, and α, β, γ, and δ are coefficients for adjusting dynamic influence;

[0184] f(x) represents the quantitative function of eloquence skill; λ1 and λ2 are weight coefficients used to adjust the influence of different skill dimensions, and x is eloquence skill;

[0185] In the complex analysis model, i is an imaginary unit, f(ζ,z) is a function of user behavior on the complex plane; C is a path around point z, representing a closed loop in time or other dimensions; ζ represents the radius of the closed loop;

[0186] In abstract algebra, i is an index variable used to identify each component in the direct sum, which varies from 1 to n; n is a positive integer, indicating the number of components in the direct sum, and is also the modulus in each modulo n integer ring, which means that each component is an integer ring modulo n; G represents the reward structure, which is a finite Abelian group used to ensure the closure and reversibility of data operations;

[0187] Furthermore, Z / nZ denotes the ring of integers modulo n, which is a set consisting of congruence classes of integers divisible by n; for example, if n=5, then Z / 5Z is the set consisting of “[0],[1],[2],[3],[4]”, where the square brackets denote congruence classes; for example, [0] contains all integers divisible by 5 (e.g., ..., -10, -5, 0, 5, 10, ...).

[0188] In this embodiment S3.5, the incentive mechanism is an important driving force for users to actively participate and improve their skills in the centralized autonomous organization. Adjusting the incentive mechanism of the centralized autonomous organization according to the eloquence performance evaluation results of the initial training data can ensure the effective allocation of resources and promote the improvement of the overall learning effect. The topological structure of the initial training data reveals information such as the skill connection and interaction pattern between users. Controlling the reward allocation algorithm to obtain this internal structure can ensure that the reward is not only based on individual performance, but also takes into account factors such as individual contribution and influence in the community, which helps to promote the healthy development of the community and enhance cooperation and interaction between users.

[0189] S3.6. In the preset evaluation system, the comprehensive quality of eloquence is calculated by multiple regression analysis based on the initial training data to obtain the comprehensive value R;

[0190] According to the time series data in the initial training data, through the calculus equation Calculate the dynamic change value of course evaluation;

[0191] Using higher-order partial derivative models The interaction and nonlinear relationship between the evaluation indicators in the initial training data are measured by partial derivatives of higher-order functions to obtain relational data;

[0192] The control evaluation system uses complex analysis to process possible complex multivariate relationships, and uses functional analysis models and abstract algebra G to optimize data structures to improve the efficiency and reliability of the evaluation system;

[0193] The evaluation data is composed of comprehensive value R, dynamic change value and relationship data;

[0194] Issue digital certificates based on the evaluation data and store them on the blockchain for verification and viewing;

[0195] Among them, the comprehensive value is:

[0196] R=β0+β1F+β2L+β3E+∈

[0197] The differential equation is:

[0198]

[0199] The high-order partial derivative model is:

[0200]

[0201] The formula used in complex analysis is:

[0202]

[0203] The functional analysis model is:

[0204] J = ∫D[f(x)]dx

[0205] Among them, R represents the final evaluation result, F, L and E represent the quantitative indicators of fluency, language richness and emotional expression respectively; β1, β2 and β3 are regression coefficients, which reflect the influence of each indicator on the final evaluation result; ∈ is a random error term, which is used to capture the random variation that cannot be explained by the model; β0 is the intercept term;

[0206] R t is the real-time evaluation index at time t, that is, the evaluation score at time t; λ is the coefficient for adjusting the sensitivity of evaluation changes. If λ is large, the evaluation score will tend to R more quickly, and if λ is small, the change of the evaluation score will be slower; R a is a reference value or a long-term average value of the evaluation, which can be fixed or expected to change over time; and, R on the right side of the differential equation t -R a Indicates the deviation between the current evaluation score and the reference value, while the left describes the rate of change of evaluation scores over time;

[0207] γ represents the effect of the interaction between fluency and language richness on evaluation;

[0208] R(z) represents the evaluation function on the complex plane. It is a complex-valued function that depends on the complex variable z. This function can be used to represent complex multivariable relationships in the evaluation system, such as the interaction between different evaluation indicators. C is the integral path on the complex plane. It is a closed curve along which the integral operation is performed. The choice of this path can be determined based on the actual situation, usually to simplify the calculation or to meet certain specific conditions. f(ζ) is the complex variable evaluation indicator function. It is a complex-valued function that depends on the complex variable ζ. This function can be used to represent a specific evaluation indicator in the evaluation system, such as user activity, learning progress, etc.

[0209] ζ is a complex variable, which is a point on the integral path C and is used to represent the variable in the integral process; z is a complex variable, which is the independent variable in the R(z) function and represents the input of the evaluation function R. In the integral expression, z is usually fixed, and ζ is the variable moving along the integral path C; dζ represents the differential element in the integral process, which is associated with the moving direction on the integral path C;

[0210] D[f(x)] represents the density function of eloquence skill data.

[0211] In this embodiment S3.6, through multiple regression analysis, the influence of multiple eloquence indicators such as fluency, language richness and emotional expression on the evaluation is comprehensively considered, which reflects the user's eloquence and expression ability;

[0212] In addition, the time series data in the initial training data contains the course evaluation information of users at different time points. The dynamic change value of the course evaluation is calculated by differential equations, which reflects the trend of course evaluation over time, which is of great significance for understanding the long-term effect of the course and predicting the future development trend of the course. There may be complex interactions and relationships between the evaluation indicators in the initial training data. Obtaining relational data through partial derivatives of higher-order functions helps to deeply understand the internal connection between the evaluation indicators and provide direction for subsequent improvement and optimization.

[0213] S3.7. The training data consists of the course evaluation data, skill data, association data, change data and rewards.

[0214] To apply the embodiments of this application, please refer to Figure 2 , Figure 2 It is a data flow diagram provided in an embodiment of the present application, which shows the general process of obtaining training data in step S3.

[0215] From a holistic perspective, this embodiment S3 obtains skill data by measuring the changes in speech skills of the initial training data. These data reflect the rise and fall of the user's speech skills; by measuring the relationship between different speech skills to obtain correlation data, the inherent connection and mutual influence between different skills are revealed, which helps to understand the overall structure and development laws of speech skills. Based on the initial training data, the spatial distribution and change trend of speech skills are simulated to obtain change data. These data show the dynamic changes of speech skills at different times and between different users, which helps to predict and plan future skill training strategies. Rewards are calculated based on the user's active participation in the course and the progress of learning, which can effectively motivate users to participate in learning more actively and improve learning motivation and effect. Combining these data to form training data can reflect the user's multi-faceted information about eloquence skill training and provide a rich data source for subsequent asset matching.

[0216] S4. In the decentralized autonomous organization, the eloquence course is divided into several assets, and the several assets are matched with several users according to the skill evaluation results and training data to obtain matching results;

[0217] Among them, the weight set of the centralized autonomous organization is obtained by adjusting the initial asset voting weights of several users in the centralized autonomous organization according to the skill evaluation results.

[0218] Step S4 of the embodiment of the present application includes S4.1 to S4.4; wherein S4.1 is the process of authenticating the current user, S4.2 is the process of obtaining the proposal content, S4.3 is the process of adjusting the DAO, and S4.4 is the process of matching assets, specifically:

[0219] S4.1. In DAO, the authentication server processes the authentication request of the current user and uses biometric authentication technology and multi-factor authentication scheme to verify the biometric identity authentication of the current user; the multi-factor authentication scheme can be a password and a mobile phone verification code, and the biometric authentication technology can be implemented through voiceprint recognition;

[0220] Use the permissions database to record each user's role (e.g., student, coach, assessor, etc.) and corresponding permission level; and use the permissions management engine to dynamically adjust the user's access rights based on the user's role and skill level to ensure that the user can only access resources open to them, such as course materials, teaching tools, assessment reports, etc.

[0221] The security monitoring center monitors all abnormal login attempts and potential security threats, responds in real time to prevent data leakage; and uses the security log system to record all key operations and security events for auditing and troubleshooting.

[0222] In this embodiment S4.1, the security of the account can be greatly improved by means of multi-factor authentication; among them, voiceprint recognition is a unique verification method related to eloquence skills, which can use the user's voice characteristics to enhance the security and personalization of the authentication process;

[0223] In addition, by logging and storing all user operations and system events, it can provide support for system security and facilitate future auditing and backtracking.

[0224] S4.2. Obtain proposals from several users;

[0225] Extract suggestions for changes to platform governance rules, user agreements, privacy policies, etc., as well as new user codes of conduct or modifications to existing codes from the proposals to obtain policy change proposals;

[0226] Extract suggestions for improving existing system functions from proposals, such as improving the friendliness of the user interface, adding new interactive functions, etc., and extract technical upgrade proposals, such as introducing more advanced algorithms or data processing technologies to improve platform performance, to obtain functional improvement proposals;

[0227] Extract new functional modules from the proposal suggestions, such as adding virtual reality (VR) elements to enhance the training experience, or introducing machine learning models to automate certain decision-making processes, to obtain new function introduction proposals;

[0228] Extract from the proposal suggestions how to use the platform funds more effectively, invest in specific technologies or market expansion, and propose specific partnerships or funding allocation strategies to obtain financial and investment proposals;

[0229] Extract proposals for activities or incentives to increase user engagement and community activity from the proposal suggestions, as well as propose plans to organize specific activities or competitions to increase user activity and loyalty, and obtain community and user engagement proposals;

[0230] Extract suggestions from the proposals to modify or optimize the governance structure of the platform, such as introducing more automated decision-making processes or adjusting the governance hierarchy to obtain proposals for governance structure adjustment;

[0231] The proposal content consists of policy change proposals, function improvement proposals, new function introduction proposals, financial and investment proposals, community and user participation proposals, and governance structure adjustment proposals.

[0232] The proposal content constructed by this embodiment S4.2 includes multiple aspects such as platform compliance, user experience, operational efficiency, innovative development and financial security, which can jointly promote the continuous and stable development of DAO.

[0233] S4.3. Use the skill evaluation results to adjust the initial asset voting weights of several users in the DAO to obtain a weight set, which ensures that more experienced and skilled users have greater influence in key decisions; and store the skill evaluation results and historical data of several users in a user skill rating database, and update the voting weights of users regularly or as needed;

[0234] Use natural language processing technology to analyze the content of proposals of several users and identify keywords and topics; match highly skilled users in related fields based on the analysis results of the proposal content, and let the matched users participate in the discussion and voting of the proposals; and implement weighted voting on key proposals to ensure that the voting results are more accurate and professional, in line with the actual needs of eloquence training and expression industries; execute the proposal content according to the discussion and voting results, and automatically adjust the governance rules of the DAO;

[0235] The real-time feedback platform allows users to submit feedback on the results of proposal execution, and analyzes the collected feedback to optimize proposals and DAO governance decisions; the currently valid governance rules and historical versions are stored in the governance rules database.

[0236] It should be noted that while ensuring that more experienced and skilled users have greater influence in key decisions through weight sets, it is necessary to set an upper limit on voting weights to avoid excessive concentration of resources, ensure that no single user or a small number of users can fully control the decision-making process, and invite third-party auditors or community representatives to participate in supervision to ensure fairness;

[0237] In addition, governance rules include:

[0238] 1. User identity and permission management rules: Determine the access rights and responsibilities of different users (such as new users, experienced speakers, administrators, etc.).

[0239] 2. Course content publishing and updating rules: specify which users can create or modify course content, and handle the approval process for related activities.

[0240] 3. Eloquence skill evaluation standards: Set standards and methods for evaluating users’ eloquence skills, such as using specific natural language processing technologies and machine learning models.

[0241] 4. Incentive distribution rules: define what behaviors or achievements can trigger incentives, as well as the type and amount of incentives.

[0242] 5. Asset fragmentation and trading rules: stipulate how to fragment eloquence courses or speech works, as well as the specific process and conditions for trading these assets.

[0243] 6. Voting and proposal system rules: including how to submit proposals, how to adjust voting weights based on users’ eloquence skill scores, and how to handle voting results.

[0244] This embodiment S4.3 adjusts the governance rules and processes according to the community's feedback and voting results, which can ensure that the governance structure can flexibly respond to changes and adapt to the needs of community growth and market development; and through a transparent real-time feedback mechanism, it enhances the transparency of governance and community participation, making the entire governance process more open and subject to community supervision.

[0245] S4.4. In DAO, register the eloquence course through the asset registration interface to make it a tradable asset;

[0246] Automatically classify assets, allowing current users to filter and invest in different types of assets based on their interests;

[0247] Divide high-value eloquence courses (such as famous teacher courses, exclusive speeches) into smaller shares to obtain a number of assets; and record the ownership of each share on the blockchain to ensure the transparency and immutability of the transaction;

[0248] While ensuring that high-priority transaction requests (such as high-value eloquence courses or transactions that need to be completed urgently) are processed first, a number of assets and a number of users are matched according to the purchase or sale requirements put forward by the user, combined with the skill assessment results and training data, to obtain a matching result; the matching of assets and users can also be achieved through asset purchase, sale and transfer in the interactive interface;

[0249] Automatically execute contract terms during transactions, including fund custody and transfer, to ensure fairness and security of transactions; and record all transaction details on the blockchain to ensure the immutability and transparency of transaction history;

[0250] Monitor market trends, provide real-time asset value analysis and investment advice, predict market trends by analyzing historical transaction data, and assist users in making more informed trading decisions.

[0251] This embodiment S4.4 can lower the investment threshold and attract a wider group of investors by dividing a single high-value asset into multiple small shares. The automatic matching mechanism can reduce user waiting time, improve transaction efficiency, reduce manual intervention, reduce error rate and fraud risk; and automatically adjust the processing priority according to the urgency and value of the transaction to ensure that important transactions can be completed first;

[0252] In addition, recording every transaction on the blockchain improves transaction transparency, allowing all parties involved to track the progress and history of transactions. By monitoring market conditions in real time, users can be provided with data-based asset evaluations, allowing users to instantly understand market dynamics such as price fluctuations and changes in supply and demand, helping them make better investment decisions.

[0253] Overall, this embodiment has the following beneficial effects:

[0254] The present invention can preliminarily understand the user's speech ability and know the user's strengths and weaknesses in speech by comprehensively evaluating the user's speech data; according to the systematic eloquence course training, the user can train his oral expression ability, logical thinking ability, psychological quality, etc. in a targeted manner, further reflecting the individual differences and advantages and disadvantages of the user in eloquence training, and the training data directly records the user's performance in the eloquence course, reflecting the user's absorption degree of the course content, adaptability to the teaching method and completion of the training task, so by observing these data, it can be known whether the course is suitable for the user's actual level and needs, revealing the preferences and adaptability of different users to different parts of the course. Since the weight set is obtained by adjusting the asset voting weights of several users in the community autonomous organization according to the skill evaluation results, the weight set can ensure that more experienced and skilled users have greater influence in key decisions on asset importance, so that user opinions can participate in community decision-making as an effective factor. Therefore, in the decentralized autonomous organization, matching several assets with several users according to the skill evaluation results and training data can fully combine community participation in decision-making, and match the corresponding assets that are currently most suitable for the user by considering the differences in user speech abilities;

[0255] In addition, by improving the DAO governance method, the problems of rigid governance structure and lack of community participation in the existing system have been solved; by introducing a weighted voting mechanism and smart contracts based on eloquence skills, the flexibility and adaptability of (voting) governance have been improved, making community management more democratic and efficient.

[0256] Embodiment 2:

[0257] See also Figure 3 , an embodiment of the present application provides a decentralized eloquence skill matching device, including a data module 10, an evaluation module 20, a training module 30 and a matching module 40;

[0258] Wherein, the data module 10 is used to obtain speech data of several users;

[0259] An evaluation module 20, used to comprehensively evaluate the speech data and obtain a skill evaluation result;

[0260] The training module 30 is used to train a number of users according to the eloquence course to obtain training data;

[0261] A matching module 40 is used to divide the eloquence course into a plurality of assets in the decentralized autonomous organization, and match the plurality of assets with a plurality of users according to the skill evaluation results and the training data to obtain a matching result;

[0262] Among them, the weight set of the centralized autonomous organization is obtained by adjusting the initial asset voting weights of several users in the centralized autonomous organization according to the skill evaluation results.

[0263] In one embodiment, the data module 10 is specifically:

[0264] Collect the speech data of several users, store the speech data in the skill archive database, and use blockchain technology to ensure the storage security and effective management of the data; among them, the speech data includes the basic information of several users (such as name, email, contact information, etc.) and eloquence-related information, and the skill archive database is used to specifically store the user's eloquence skill certification documents, such as certification, speech videos, etc.

[0265] In the preset DAO (Decentralized Autonomous Organization), a unique identity is created based on the basic information of the current user; the current user refers to the user who needs to obtain resource matching permissions to implement this solution.

[0266] In one embodiment, the evaluation module 20 includes an emotion unit, a knowledge unit, and a skill unit; wherein the emotion unit, the knowledge unit, and the skill unit are the process of calculating the skill evaluation result, specifically:

[0267] The sentiment unit is used to establish fuzzy sets based on the expressiveness index, pitch variation, language richness and knowledge depth as category standards; and the fuzzy set allows variables to exist in a way of "belonging" or "not belonging" to a certain degree. The set can be divided into different fuzzy levels according to actual needs, such as "very weak", "weak", "medium", "strong" and "very strong", etc.; and define a membership function for the fuzzy set, which describes the degree to which the variable value belongs to the corresponding category of the set;

[0268] The sentiment unit is also used to convert discrete or approximately discrete data in the speech data into fuzzy values ​​by mapping the speech data to corresponding membership degrees in the fuzzy set to obtain a fuzzy value set;

[0269] The sentiment unit is also used to defuzzify the fuzzy value set by the maximum membership method to obtain a continuous data set;

[0270] The emotion unit is also used to calculate the expressiveness index EI (t) based on the emotional intensity and change rate of the continuous data set with the help of the skill analysis engine, using the expressiveness index dynamic model EI (t); the skill analysis engine has built-in natural language processing and machine learning algorithms;

[0271] The emotion unit is also used to calculate the pitch change degree PT according to the pitch change amplitude and flexibility of the continuous data set;

[0272] Knowledge unit, used to calculate the language richness LR based on the vocabulary usage frequency and sentence structure of the continuous data set;

[0273] The knowledge unit is also used to calculate the knowledge depth KD based on the coverage and usage accuracy of knowledge points in the continuous data set;

[0274] Skill unit, which is used to form the skill evaluation result from the expression index EI, pitch change PT, language richness LR and knowledge depth KD, and record the skill evaluation result in the user's blockchain profile;

[0275] Among them, the dynamic model of the expressiveness index is:

[0276]

[0277] The pitch variation is:

[0278]

[0279] The language richness is:

[0280]

[0281] The depth of knowledge is:

[0282]

[0283] Among them, E(t) represents the intensity of emotion at time t, α, β and γ are coefficients for adjusting the dynamic influence of emotion;

[0284] represents the intensity of the tone at time t and frequency f; represents the second-order partial derivative with respect to time, which is used to measure the acceleration of pitch change;

[0285] V i represents the usage frequency of the i-th category of words, λ i is the weight factor of the i-th category of vocabulary, and N is the total number of categories of vocabulary; represents the gradient operator used to find the vocabulary usage strategy that maximizes language richness;

[0286] K(t) represents the coverage of knowledge points at time t, A(t) represents the accuracy of the corresponding knowledge points at time t, and T is the total duration of the speech.

[0287] In the emotion unit, knowledge unit and skill unit of this embodiment, the expressiveness index can measure the clarity and effectiveness of the information conveyed by the user; the pitch variation measures the variation and flexibility of the user's voice. A higher pitch variation may indicate that the user can attract the audience's attention through the fluctuation of the voice and enhance the effect of expression; the language richness measures the diversity and complexity of the user's language through the frequency of vocabulary use; the knowledge depth can evaluate the user's knowledge mastery in a certain field or topic, reflecting the user's understanding and professional quality in the field; therefore, the evaluation results of these four aspects can form a comprehensive evaluation of the user's eloquence or speech skills, reflecting the user's comprehensive speech ability;

[0288] Furthermore, this embodiment introduces calculus equations and partial derivative formulas of higher-order functions to describe and optimize continuous changes in the speech evaluation process, which can effectively increase the accuracy and flexibility of the model in processing continuous data streams and dynamic adjustments. By providing a transparent skill evaluation process, users can view and verify their evaluation history and results at any time.

[0289] In one embodiment, the training module 30 includes an optimization unit, a monitoring unit, a curriculum unit, a training unit, a variation unit, a mapping unit, a distribution unit, a reward unit, an incentive unit, a structure unit, a consistency unit, an evaluation unit, and a synthesis unit;

[0290] Among them, the optimization unit is the process of establishing the optimization problem of course adjustment, the monitoring unit is the process of continuously monitoring the eloquence skills to obtain the eloquence skill change value, the course unit is the process of establishing the eloquence course based on the optimization problem and the eloquence skill change value, the training unit, the change unit, the mapping unit and the distribution unit are the processes of calculating the skill data, the associated data and the change data, the reward unit, the incentive unit, the structure unit and the consistency unit are the processes of calculating the reward and making preliminary adjustments to the mechanism, the evaluation unit is the process of calculating the evaluation data, and the comprehensive unit is the process of generating the training data based on the evaluation data, the skill data, the associated data, the change data and the reward, specifically:

[0291] Optimization units are used to define the ideal state that you want to achieve at the end of a pre-set course;

[0292] The optimization unit is also used to establish the terminal cost function g(c(T)) by quantifying the gap between the state at the end of the preset course and the ideal state;

[0293] The optimization unit is further used to establish a loss function L(c(t),t) according to the inadaptability loss of a number of users to a course configuration c(t) in a preset course;

[0294] The optimization unit is also used to establish an optimization problem F about course adjustment according to the terminal cost function g(c(T)), the course configuration c(t) and the loss function L(c(t),t);

[0295] Among them, the optimization problem of course adjustment is:

[0296]

[0297] Among them, c(t) represents the course configuration at time t; L(c(t), t) is the loss function, which reflects the user's inadaptability to the current course configuration; g(c(T)) is the terminal cost function, which reflects the state at the end of the course; T is the total duration of the preset course.

[0298] The optimization unit of this embodiment can clearly see the degree of deviation between the current course configuration and the ideal state by quantifying the gap between the state at the end of the course and the ideal state. The terminal cost function is a mathematical tool for evaluating the gap between the state at the end of the course and the ideal state. Through this function, the effect of the current course configuration can be more accurately described, providing quantitative guidance for course adjustment. A loss function is established based on the inadaptability loss of several users to the course configuration, which can provide an in-depth understanding of the user's feedback and opinions on the current course configuration. An optimization problem for course adjustment is constructed by combining the course configuration, the loss function and the terminal cost function. This optimization problem is actually a mathematical model. By solving the optimization problem, a set of optimal course configuration solutions can be obtained. These solutions can guide course designers to improve the current course configuration to improve the teaching effect of the course and user satisfaction, and can meet the needs of real-time adjustment of course content.

[0299] A monitoring unit, used to simulate the skill changes of several users according to a preset partial differential equation to obtain a change value of eloquence skill;

[0300] The partial differential equation is:

[0301]

[0302] Among them, s(x,t) represents the skill level at position x and time t, D is the diffusion coefficient, λ is the attenuation coefficient, and f(x,t) is the parameter input by external training.

[0303] The monitoring unit of this embodiment can accurately simulate the changes of the user's eloquence skills over time or other factors through the preset partial differential equation, provide strong data support for subsequent course adjustments, and meet the needs of real-time monitoring and updating of the user's eloquence skills.

[0304] The course unit is used to map the user performance in the speech data to the evaluation space of the user performance in the algebraic structure C, and to establish the eloquence course by combining the optimization problem and the eloquence skill change value; wherein the algebraic structure C is the mathematical structure configured by the preset course;

[0305] The algebraic structure is:

[0306]

[0307] The process of mapping user performance to evaluation space can be expressed as:

[0308] E:X→Y

[0309] Among them, C represents the algebraic structure of the entire course data, which is a set that contains the data of all course modules; C i It is the data representation of a single course module, such as a chapter, a lesson, or a specific skill training unit; i is the ordinal number of the course module, and n is the total number of course modules; The direct sum of the data structure is used to ensure the independence and overall consistency between data components. In this embodiment, the direct sum ensures that each course module C i While functioning as independent units, they together form a coherent structure for the entire Course C.

[0310] X is the set of user performances, Y is the space of user performance evaluation results; E represents a continuous mapping from X to Y, which is used to ensure the consistency and fairness of the evaluation.

[0311] The course unit of this embodiment can determine the goal of course adjustment by analyzing the causes of inadaptability, that is, how to reduce or eliminate these inadaptabilities by changing the course content, teaching methods or progress. By simulating the changes in users' skills, it is possible to predict or evaluate the impact of different teaching methods, course content or progress on users' eloquence skills, which helps to understand which teaching method or content adjustment is more likely to promote user skill improvement. Mapping eloquence data to the evaluation space of user performance in an algebraic structure can convert qualitative or complex data into a form that can be quantified for analysis and comparison, thereby providing more accurate guidance for course adjustment; this way of establishing eloquence courses can ensure that the course content and teaching methods can meet the needs of users to the greatest extent possible;

[0312] And, in abstract algebra, each component C iIt has its own algebraic properties (such as the structure of a group, ring or field), so that data operations (such as adding, updating or deleting information) can be performed while maintaining structural and mathematical properties; this structured data processing method helps to ensure the correctness of operations and the integrity of data; the use of abstract algebraic structures can systematically manage course data, and the mathematical rules in it can ensure that all operations do not violate the defined data rules. For example, if course information is connected through a specific functional relationship (such as mapping or transformation), when the data is updated or modified, the related functions will also be updated accordingly, thus ensuring the consistency of the data.

[0313] A training unit, used to train a number of users using the eloquence course to obtain initial training data; wherein the initial training data includes the feedback of the users on the course and the evaluation of the teacher;

[0314] The change unit is used to construct a Hilbert Space suitable for eloquence skill evaluation; according to the functional model Φ(f), the initial training data is converted into elements that can be represented in the Hilbert Space by feature extraction and encoding; the change of the initial training data in speech skills is measured by the inner product operation in the Hilbert Space to obtain the skill data;

[0315] A mapping unit, used for measuring the complex associations between different speech skills in the initial training data through a complex variable mapping G(z), and measuring the continuity and changes between different speech skills in the initial training data through a continuous mapping in topology, to obtain association data;

[0316] A distribution unit is used to simulate the spatial distribution and change trend of speech skills based on the initial training data and the geometric surface model using the curve and surface theory in geometry to obtain change data;

[0317] Among them, the functional model is:

[0318] Φ(f)=∫ D (K(x,f(x))dx)

[0319] The complex variable mapping can be expressed as:

[0320]

[0321] A continuous map in topology can be represented as:

[0322] ψ:X→Y, continuous Opening ψ -1 (V) is also an open set in X

[0323] The geometric surface model is:

[0324] S(t,θ)=(tcosθ,tsinθ,sinf(t))

[0325] Where f(x) represents a specific skill performance in the speech process, such as pitch or speaking speed; K(x,y) is a kernel function used to deal with nonlinear relationships in speech data; D is the time region of the speech duration;

[0326] f(ζ) represents a complex analytical function of eloquence performance, C is a closed path around the observation point z, and i is an imaginary unit;

[0327] X and Y represent the topological space of skill levels before and after speech, respectively;

[0328] t and θ represent the variation parameters of time and pitch, and f(t) represents the skill variation function related to time.

[0329] The training unit, change unit, mapping unit and distribution unit of this embodiment use the complex analysis method to deeply understand the internal structure of speech skill changes, especially the data association when dealing with periodic and step changes, providing strong data support for subsequent asset matching. Since the initial training data includes feedback from several users on the course and evaluation of the teacher, it can also help other users to select course assets.

[0330] The reward unit is used to calculate the impact of user behavior changes on eloquence improvement based on the course activity and learning progress of several users in the initial training data using the incentive function model R(t) to obtain rewards;

[0331] The incentive unit is used to construct a functional space based on the initial training data and use the eloquence performance functional J[f] to evaluate the eloquence performance and obtain the evaluation results; and use the evaluation results and the optimization of the functional to adjust the incentive mechanism of the DAO;

[0332] The structural unit is used to process the complex patterns of user behaviors in the initial training data through the complex analysis model F(z), obtain the topological structure, and control the DAO's reward distribution algorithm to obtain and grasp the internal structure of the initial training data according to the topological structure;

[0333] Consistency unit, which is used to ensure the integrity and consistency of data in the incentive mechanism through the preset abstract algebraic G structure; using smart contracts to ensure the transparency and fairness of the incentive mechanism, making all reward records public and traceable;

[0334] Among them, the activation function model is:

[0335]

[0336] The eloquence performance functional is:

[0337]

[0338] The complex analysis model is:

[0339]

[0340] The abstract algebra is:

[0341]

[0342] Where A(t) is the user's activity at time t, P(t) is the learning progress, and α, β, γ, and δ are coefficients for adjusting dynamic influence;

[0343] f(x) represents the quantitative function of eloquence skill; λ1 and λ2 are weight coefficients used to adjust the influence of different skill dimensions, and x is eloquence skill;

[0344] In the complex analysis model, i is an imaginary unit, f(ζ,z) is a function of user behavior on the complex plane; C is a path around point z, representing a closed loop in time or other dimensions; ζ represents the radius of the closed loop;

[0345] In abstract algebra, i is an index variable used to identify each component in the direct sum, which varies from 1 to n; n is a positive integer, indicating the number of components in the direct sum, and is also the modulus in each modulo n integer ring, which means that each component is an integer ring modulo n; G represents the reward structure, which is a finite Abelian group used to ensure the closure and reversibility of data operations;

[0346] Furthermore, Z / nZ denotes the ring of integers modulo n, which is a set consisting of congruence classes of integers divisible by n; for example, if n=5, then Z / 5Z is the set consisting of “[0],[1],[2],[3],[4]”, where the square brackets denote congruence classes; for example, [0] contains all integers divisible by 5 (e.g., ..., -10, -5, 0, 5, 10, ...).

[0347] In the reward unit, incentive unit, structure unit and consistency unit of this embodiment, the incentive mechanism is an important driving force for users to actively participate and improve their skills in the centralized autonomous organization. Adjusting the incentive mechanism of the centralized autonomous organization according to the eloquence performance evaluation results of the initial training data can ensure the effective allocation of resources and promote the improvement of the overall learning effect. The topological structure of the initial training data reveals information such as the skill connection and interaction mode between users. Controlling the reward allocation algorithm to obtain this internal structure can ensure that the reward is not only based on individual performance, but also takes into account factors such as individual contribution and influence in the community, which helps to promote the healthy development of the community and enhance cooperation and interaction between users.

[0348] The evaluation unit is used to calculate the comprehensive quality of eloquence in a preset evaluation system according to the initial training data by means of multiple regression analysis to obtain a comprehensive value R;

[0349] The evaluation unit is also used to evaluate the time series data in the initial training data through calculus equations Calculate the dynamic change value of course evaluation;

[0350] Evaluation unit, also used to use higher-order partial derivative models The interaction and nonlinear relationship between the evaluation indicators in the initial training data are measured by partial derivatives of higher-order functions to obtain relational data;

[0351] The evaluation unit is also used to control the evaluation system to process possible complex multivariate relationships through complex analysis, and to optimize the data structure using functional analysis models and abstract algebra G to improve the efficiency and reliability of the evaluation system;

[0352] The evaluation unit is also used to form evaluation data from the comprehensive value R, the dynamic change value and the relationship data;

[0353] The evaluation unit is also used to issue a digital certificate based on the evaluation data and store the digital certificate on the blockchain for verification and viewing;

[0354] Among them, the comprehensive value is:

[0355] R=β0+β1F+β2L+β3E+∈

[0356] The differential equation is:

[0357]

[0358] The high-order partial derivative model is:

[0359]

[0360] The formula used in complex analysis is:

[0361]

[0362] The functional analysis model is:

[0363] J = ∫D[f(x)]dx

[0364] Among them, R represents the final evaluation result, F, L and E represent the quantitative indicators of fluency, language richness and emotional expression respectively; β1, β2 and β3 are regression coefficients, which reflect the influence of each indicator on the final evaluation result; ∈ is a random error term, which is used to capture the random variation that cannot be explained by the model; β0 is the intercept term;

[0365] R t is the real-time evaluation index at time t, that is, the evaluation score at time t; λ is the coefficient for adjusting the sensitivity of evaluation changes. If λ is large, the evaluation score will tend to R more quickly, and if λ is small, the change of the evaluation score will be slower; R a is a reference value or a long-term average value of the evaluation, which can be fixed or expected to change over time; and, R on the right side of the differential equation t -R a Indicates the deviation between the current evaluation score and the reference value, while the left describes the rate of change of evaluation scores over time;

[0366] γ represents the effect of the interaction between fluency and language richness on evaluation;

[0367] R(z) represents the evaluation function on the complex plane. It is a complex-valued function that depends on the complex variable z. This function can be used to represent complex multivariable relationships in the evaluation system, such as the interaction between different evaluation indicators. C is the integral path on the complex plane. It is a closed curve along which the integral operation is performed. The choice of this path can be determined based on the actual situation, usually to simplify the calculation or to meet certain specific conditions. f(ζ) is the complex variable evaluation indicator function. It is a complex-valued function that depends on the complex variable ζ. This function can be used to represent a specific evaluation indicator in the evaluation system, such as user activity, learning progress, etc.

[0368] ζ is a complex variable, which is a point on the integral path C and is used to represent the variable in the integral process; z is a complex variable, which is the independent variable in the R(z) function and represents the input of the evaluation function R. In the integral expression, z is usually fixed, and ζ is the variable moving along the integral path C; dζ represents the differential element in the integral process, which is associated with the moving direction on the integral path C;

[0369] D[f(x)] represents the density function of eloquence skill data.

[0370] The evaluation unit of this embodiment uses multiple regression analysis to comprehensively consider the impact of multiple eloquence indicators such as fluency, language richness and emotional expression on the evaluation, reflecting the user's eloquence and expression ability;

[0371] In addition, the time series data in the initial training data contains the course evaluation information of users at different time points. The dynamic change value of the course evaluation is calculated by differential equations, which reflects the trend of course evaluation over time, which is of great significance for understanding the long-term effect of the course and predicting the future development trend of the course. There may be complex interactions and relationships between the evaluation indicators in the initial training data. Obtaining relational data through partial derivatives of higher-order functions helps to deeply understand the internal connection between the evaluation indicators and provide direction for subsequent improvement and optimization.

[0372] A comprehensive unit is used to form training data consisting of evaluation data, skill data, association data, change data and rewards of the course.

[0373] To apply the embodiments of this application, please refer to Figure 2 , Figure 2 It is a data flow diagram provided in an embodiment of the present application, which shows the general process of obtaining training data in step S3.

[0374] From a holistic perspective, the training module 30 of this embodiment obtains skill data by measuring the changes in speech skills of the initial training data. These data reflect the user's improvement or decline in speech skills; by measuring the relationship between different speech skills to obtain correlation data, the inherent connection and mutual influence between different skills are revealed, which helps to understand the overall structure and development laws of speech skills. Based on the initial training data, the spatial distribution and change trend of speech skills are simulated to obtain change data. These data show the dynamic changes of speech skills at different times and between different users, which helps to predict and plan future skill training strategies. Rewards are calculated based on the user's activeness in participating in the course and the learning progress, which can effectively motivate users to participate in learning more actively and improve learning motivation and results. Combining these data to form training data can reflect the user's multi-faceted information about eloquence skill training and provide a rich data source for subsequent asset matching.

[0375] In one embodiment, the matching module 40 includes a verification unit, a proposal unit, an adjustment unit and a matching unit; wherein the verification unit is a process of authenticating the current user, the proposal unit is a process of obtaining the proposal content, the adjustment unit is a process of adjusting the DAO, and the matching unit is a process of matching assets, specifically:

[0376] The verification unit is used to process the authentication request of the current user through the authentication server in the DAO, and use the biometric authentication technology and the multi-factor authentication scheme to verify the biometric identity authentication of the current user; wherein the multi-factor authentication scheme can be a password and a mobile phone verification code, and the biometric authentication technology can be implemented by voiceprint recognition;

[0377] The verification unit is also used to use the permission database to record each user's role (such as student, coach, assessor, etc.) and the corresponding permission level; and according to the user's role and skill level, the permission management engine is used to dynamically adjust the user's access rights to ensure that the user can only access resources open to him, such as course materials, teaching tools, evaluation reports, etc.;

[0378] The verification unit is also used to monitor all abnormal login attempts and potential security threats through the security monitoring center, respond in real time to prevent data leakage; and use the security log system to record all key operations and security events for auditing and troubleshooting.

[0379] The verification unit of this embodiment can greatly improve the security of the account through multi-factor authentication. Among them, voiceprint recognition is a unique verification method related to eloquence skills, which can use the user's voice characteristics to enhance the security and personalization of the authentication process;

[0380] In addition, by logging and storing all user operations and system events, it can provide support for system security and facilitate future auditing and backtracking.

[0381] Proposal unit, used to obtain proposal suggestions from several users;

[0382] The proposal unit is also used to extract proposals for changes to platform governance rules, user agreements, privacy policies, etc., as well as new user codes of conduct or modifications to existing codes from proposal suggestions to obtain policy change proposals;

[0383] The proposal unit is also used to extract suggestions for improving existing system functions from proposals, such as improving the friendliness of the user interface, adding new interactive functions, etc., and extract technical upgrade proposals, such as introducing more advanced algorithms or data processing technologies to improve platform performance, and obtain functional improvement proposals;

[0384] The proposal unit is also used to extract new functional modules from the proposal suggestions, such as adding virtual reality (VR) elements to enhance the training experience, or introducing machine learning models to automate certain decision-making processes, to obtain new function introduction proposals;

[0385] The proposal unit is also used to extract suggestions on how to use platform funds more effectively, invest in specific technologies or market expansion, and propose specific partnerships or funding allocation strategies from proposal suggestions to obtain financial and investment proposals;

[0386] The proposal unit is also used to extract proposals for activities or incentive mechanisms aimed at improving user participation and community activity from proposal suggestions, as well as to propose plans to organize specific activities or competitions to increase user activity and loyalty, and obtain community and user participation proposals;

[0387] The proposal unit is also used to extract suggestions from the proposal suggestions to modify or optimize the governance structure of the platform, such as introducing more automated decision-making processes or adjusting the governance hierarchy to obtain proposals for governance structure adjustment;

[0388] The proposal unit is also used to form proposal content consisting of policy change proposals, function improvement proposals, new function introduction proposals, financial and investment proposals, community and user participation proposals, and governance structure adjustment proposals.

[0389] The proposal content constructed by the proposal unit of this embodiment includes multiple aspects such as platform compliance, user experience, operational efficiency, innovative development and financial security, which can jointly promote the continuous and stable development of DAO.

[0390] An adjustment unit is used to use the skill evaluation results to adjust the initial asset voting weights of several users in the DAO to obtain a weight set, which ensures that more experienced and skilled users have greater influence in key decisions; and store the skill evaluation results and historical data of several users in a user skill rating database, and update the voting weights of users regularly or on demand;

[0391] The adjustment unit is also used to analyze the proposal content of several users using natural language processing technology to identify keywords and topics; match highly skilled users in related fields based on the analysis results of the proposal content, and let the matched users participate in the discussion and voting of the proposal; and implement weighted voting on key proposals to ensure that the voting results are more accurate and professional, in line with the actual needs of eloquence training and expression industries; execute the proposal content according to the discussion and voting results, and automatically adjust the governance rules of the DAO;

[0392] The adjustment unit is also used to allow users to submit feedback on the results of proposal execution through the real-time feedback platform, and analyze the collected feedback to optimize the proposal and DAO governance decisions; the current effective governance rules and historical versions are stored in the governance rules database.

[0393] It should be noted that while ensuring that more experienced and skilled users have greater influence in key decisions through weight sets, it is necessary to set an upper limit on voting weights to avoid excessive concentration of resources, ensure that no single user or a small number of users can fully control the decision-making process, and invite third-party auditors or community representatives to participate in supervision to ensure fairness;

[0394] In addition, governance rules include:

[0395] 1. User identity and permission management rules: Determine the access rights and responsibilities of different users (such as new users, experienced speakers, administrators, etc.).

[0396] 2. Course content publishing and updating rules: specify which users can create or modify course content, and handle the approval process for related activities.

[0397] 3. Eloquence skill evaluation standards: Set standards and methods for evaluating users’ eloquence skills, such as using specific natural language processing technologies and machine learning models.

[0398] 4. Incentive distribution rules: define what behaviors or achievements can trigger incentives, as well as the type and amount of incentives.

[0399] 5. Asset fragmentation and trading rules: stipulate how to fragment eloquence courses or speech works, as well as the specific process and conditions for trading these assets.

[0400] 6. Voting and proposal system rules: including how to submit proposals, how to adjust voting weights based on users’ eloquence skill scores, and how to handle voting results.

[0401] The adjustment unit of this embodiment adjusts the governance rules and processes according to the community's feedback and voting results, which can ensure that the governance structure can flexibly respond to changes and adapt to the needs of community growth and market development; and through a transparent real-time feedback mechanism, it enhances the transparency of governance and community participation, making the entire governance process more open and subject to community supervision.

[0402] The matching unit is used to register the eloquence course in the DAO through the asset registration interface, making it a tradable asset;

[0403] The matching unit is also used to automatically classify assets, allowing current users to filter and invest in different types of assets according to their interests;

[0404] Matching units are also used to divide high-value eloquence courses (such as famous teacher courses and exclusive speeches) into smaller shares to obtain a number of assets; and record the ownership of each share on the blockchain to ensure the transparency and immutability of the transaction;

[0405] The matching unit is also used to match a number of assets with a number of users based on the purchase or sale requirements put forward by the user, in combination with the skill assessment results and training data, to obtain a matching result, while ensuring that high-priority transaction requests (such as high-value eloquence courses or transactions that need to be completed urgently) are processed first; the matching of assets and users can also be achieved through asset purchase, sale and transfer in the interactive interface;

[0406] The matching unit is also used to automatically execute the contract terms in the transaction process, including fund custody and transfer, to ensure the fairness and security of the transaction; and record all transaction details on the blockchain to ensure the immutability and transparency of the transaction history;

[0407] The matching unit is also used to monitor market dynamics, provide real-time asset value analysis and investment advice, predict market trends by analyzing historical transaction data, and assist users in making more informed trading decisions.

[0408] The matching unit of this embodiment can lower the investment threshold and attract a wider group of investors by dividing a single high-value asset into multiple small shares. The automatic matching mechanism can reduce user waiting time, improve transaction efficiency, reduce manual intervention, reduce error rate and fraud risk; and automatically adjust the processing priority according to the urgency and value of the transaction to ensure that important transactions can be completed first;

[0409] In addition, recording every transaction on the blockchain improves transaction transparency, allowing all parties involved to track the progress and history of transactions. By monitoring market conditions in real time, users can be provided with data-based asset evaluations, allowing users to instantly understand market dynamics such as price fluctuations and changes in supply and demand, helping them make better investment decisions.

[0410] Overall, this embodiment has the following beneficial effects:

[0411] The present invention can preliminarily understand the user's speech ability and know the user's strengths and weaknesses in speech by comprehensively evaluating the user's speech data; according to the systematic eloquence course training, the user can train his oral expression ability, logical thinking ability, psychological quality, etc. in a targeted manner, further reflecting the individual differences and advantages and disadvantages of the user in eloquence training, and the training data directly records the user's performance in the eloquence course, reflecting the user's absorption degree of the course content, adaptability to the teaching method and completion of the training task, so by observing these data, it can be known whether the course is suitable for the user's actual level and needs, revealing the preferences and adaptability of different users to different parts of the course. Since the weight set is obtained by adjusting the asset voting weights of several users in the community autonomous organization according to the skill evaluation results, the weight set can ensure that more experienced and skilled users have greater influence in key decisions on asset importance, so that user opinions can participate in community decision-making as an effective factor. Therefore, in the decentralized autonomous organization, matching several assets with several users according to the skill evaluation results and training data can fully combine community participation in decision-making, and match the corresponding assets that are currently most suitable for the user by considering the differences in user speech abilities;

[0412] In addition, by improving the DAO governance method, the problems of rigid governance structure and lack of community participation in the existing system have been solved; by introducing a weighted voting mechanism and smart contracts based on eloquence skills, the flexibility and adaptability of (voting) governance have been improved, making community management more democratic and efficient.

[0413] Embodiment three:

[0414] The embodiment of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium includes a stored computer program, wherein when the computer program is running, the device where the computer-readable storage medium is located is controlled to execute the decentralized eloquence skill matching method;

[0415] Wherein, the decentralized eloquence skill matching method, if implemented in the form of a software functional unit and used as an independent product, can be stored in a computer-readable storage medium. Based on such an understanding, the present invention implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. Wherein, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form, etc. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, U disk, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electric carrier signal, telecommunication signal and software distribution medium, etc.

[0416] The above are preferred embodiments of the present invention. It should be noted that, for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications are also considered to be within the scope of protection of the present invention.

Claims

1. A decentralized eloquence skill matching method, characterized in that: include: Get speech data of several users; Performing a comprehensive evaluation on the speech data to obtain a skill evaluation result; Training the plurality of users according to the eloquence course to obtain training data; In a decentralized autonomous organization, the eloquence course is divided into a plurality of assets, and the plurality of assets are matched with the plurality of users according to the skill evaluation result and the training data to obtain a matching result; The weight set of the decentralized autonomous organization is obtained by adjusting the initial asset voting weights of the plurality of users in the decentralized autonomous organization according to the skill evaluation results; the weight set is used to ensure that more experienced and skilled users have greater influence in key decisions; The eloquence courses are specifically: According to the adaptability of the plurality of users to the preset courses, an optimization problem regarding course adjustment is established, specifically: defining the ideal state desired to be achieved at the end of the preset course; establishing a terminal cost function by quantifying the gap between the state at the end of the preset course and the ideal state; establishing a loss function according to the inadaptability loss of the plurality of users to the course configuration in the preset course; establishing the optimization problem regarding course adjustment according to the terminal cost function, the course configuration and the loss function; By simulating the skill changes of the plurality of users, a change value of eloquence skill is obtained; The speech data is mapped to an evaluation space of user performance in an algebraic structure, and the eloquence course is established by combining the optimization problem and the eloquence skill change value; wherein the algebraic structure is a mathematical structure configured by the preset course.

2. A decentralized eloquence skill matching method as claimed in claim 1, characterized in that: A comprehensive evaluation is performed on the speech data to obtain a skill evaluation result, specifically: According to the emotion intensity and tone intensity of the speech data at different times, respectively calculating the expressiveness index and the tone variation degree; Calculating language richness and knowledge depth respectively based on the vocabulary usage frequency and knowledge point usage accuracy of the speech data; The skill evaluation result is composed of the expressiveness index, the pitch variation, the language richness and the knowledge depth.

3. A decentralized eloquence skill matching method as claimed in claim 1, characterized in that: By simulating the skill changes of the several users, the eloquence skill change value is obtained, which is specifically: Simulating the skill changes of the plurality of users according to a preset partial differential equation to obtain the eloquence skill change value; Wherein, the partial differential equation is: Among them, s(x,t) represents the skill level at position x and time t, D is the diffusion coefficient, λ is the attenuation coefficient, and f(x,t) is the parameter input by external training.

4. A decentralized eloquence skill matching method as claimed in claim 1, characterized in that: The users are trained according to the eloquence course to obtain training data, specifically: Using the eloquence course to train the plurality of users to obtain initial training data; Measuring the speech skill change of the initial training data to obtain skill data; Measuring the association between different speech skills in the initial training data by complex variable mapping and continuous mapping to obtain association data; Based on the initial training data, the spatial distribution and change trend of speech skills are simulated to obtain change data, specifically: based on the initial training data, according to the geometric surface model, the curve and surface theory in geometry is used to simulate the spatial distribution and change trend of speech skills to obtain change data; Calculate rewards based on the course activity and learning progress of the several users in the initial training data; The training data is composed of the course evaluation data, the skill data, the association data, the change data and the reward; The evaluation data of the course is specifically: According to the initial training data, the comprehensive quality of eloquence is calculated by means of multiple regression analysis to obtain a comprehensive value; According to the time series data in the initial training data, a dynamic change value of the course evaluation is calculated by a differential equation; Measuring the effects and relationships between the evaluation indicators in the initial training data by using partial derivatives of higher-order functions to obtain relationship data; The evaluation data is composed of the comprehensive value, the dynamic change value and the relationship data.

5. A decentralized eloquence skill matching method as claimed in claim 4, characterized in that: After calculating the reward according to the course activity and learning progress of the plurality of users in the initial training data, the method further includes: Evaluating eloquence performance based on the initial training data, and using the evaluation results to adjust the incentive mechanism of the decentralized autonomous organization; According to the topological structure of the initial training data, controlling the reward allocation algorithm of the decentralized autonomous organization to obtain the internal structure of the initial training data; The integrity and consistency of the data in the incentive mechanism are ensured by a preset abstract algebraic structure.

6. A decentralized eloquence skill matching device, characterized in that: Includes data module, evaluation module, training module and matching module; Wherein, the data module is used to obtain speech data of several users; The evaluation module is used to comprehensively evaluate the speech data to obtain a skill evaluation result; The training module is used to train the plurality of users according to the eloquence course to obtain training data; The matching module is used to divide the eloquence course into a plurality of assets in the decentralized autonomous organization, and match the plurality of assets with the plurality of users according to the skill evaluation result and the training data to obtain a matching result; The weight set of the decentralized autonomous organization is obtained by adjusting the initial asset voting weights of the plurality of users in the decentralized autonomous organization according to the skill evaluation results; the weight set is used to ensure that more experienced and skilled users have greater influence in key decisions; The eloquence courses are specifically: According to the adaptability of the plurality of users to the preset courses, an optimization problem regarding course adjustment is established, specifically: defining the ideal state desired to be achieved at the end of the preset course; establishing a terminal cost function by quantifying the gap between the state at the end of the preset course and the ideal state; establishing a loss function according to the inadaptability loss of the plurality of users to the course configuration in the preset course; establishing the optimization problem regarding course adjustment according to the terminal cost function, the course configuration and the loss function; By simulating the skill changes of the plurality of users, a change value of eloquence skill is obtained; The speech data is mapped to an evaluation space of user performance in an algebraic structure, and the eloquence course is established by combining the optimization problem and the eloquence skill change value; wherein the algebraic structure is a mathematical structure configured by the preset course.

7. A decentralized eloquence skill matching device as claimed in claim 6, characterized in that: The evaluation module includes an emotion unit, a knowledge unit and a skill unit; The emotion unit is used to calculate the expression index and the pitch change degree according to the emotion intensity and pitch intensity of the speech data at different times; The knowledge unit is used to calculate the language richness and knowledge depth respectively according to the vocabulary usage frequency and knowledge point usage accuracy of the speech data; The skill unit is used to form the skill evaluation result by the expressiveness index, the pitch variation, the language richness and the knowledge depth.

8. A decentralized eloquence skill matching device as claimed in claim 6, characterized in that: By simulating the skill changes of the several users, the eloquence skill change value is obtained, which is specifically: Simulating the skill changes of the plurality of users according to a preset partial differential equation to obtain the eloquence skill change value; Wherein, the partial differential equation is: Among them, s(x,t) represents the skill level at position x and time t, D is the diffusion coefficient, λ is the attenuation coefficient, and f(x,t) is the parameter input by external training.

9. A decentralized eloquence skill matching device as claimed in claim 6, characterized in that: The training module includes a training unit, a variation unit, a mapping unit, a distribution unit, a reward unit and a comprehensive unit; The training unit is used to train the plurality of users using the eloquence course to obtain initial training data; The change unit is used to measure the speech skill change of the initial training data to obtain skill data; The mapping unit is used to measure the association between different speech skills in the initial training data through complex variable mapping and continuous mapping to obtain association data; The distribution unit is used to simulate the spatial distribution and change trend of speech skills based on the initial training data to obtain change data, specifically: based on the initial training data, according to the geometric surface model, using the curve and surface theory in geometry to simulate the spatial distribution and change trend of speech skills to obtain change data; The reward unit is used to calculate a reward according to the course activity and learning progress of the plurality of users in the initial training data; The integration unit is used to form the training data from the course evaluation data, the skill data, the association data, the change data and the reward; The evaluation data of the course is specifically: According to the initial training data, the comprehensive quality of eloquence is calculated by means of multiple regression analysis to obtain a comprehensive value; According to the time series data in the initial training data, a dynamic change value of the course evaluation is calculated by a differential equation; Measuring the effects and relationships between the evaluation indicators in the initial training data by using partial derivatives of higher-order functions to obtain relationship data; The evaluation data is composed of the comprehensive value, the dynamic change value and the relationship data.

10. A decentralized eloquence skill matching device as claimed in claim 9, characterized in that: After the reward unit, it also includes an incentive unit, a structure unit and a consistency unit; The incentive unit is used to evaluate the eloquence performance according to the initial training data, and use the evaluation result to adjust the incentive mechanism of the decentralized autonomous organization; The structural unit is used to control the reward allocation algorithm of the decentralized autonomous organization to obtain the internal structure of the initial training data according to the topological structure of the initial training data; The consistency unit is used to ensure the integrity and consistency of the data in the incentive mechanism through a preset abstract algebraic structure.

11. A storage medium, characterized in that: The storage medium stores a computer program, which is called and executed by a computer to implement any one of the decentralized eloquence skill matching methods of claims 1 to 5 above.

Images