Environmental art design course teaching quality monitoring method based on online data

Abstract

The invention provides an environmental art design course teaching quality monitoring method based on online data, and relates to the technical field of education, and the method comprises the steps: collecting the teaching data of an environmental art design course in real time through an online teaching platform, including the student participation degree, the homework submission condition, the online test score, the classroom interaction record and the teacher teaching feedback; and according to the teaching data, through clustering analysis and association rule analysis, constructing a teaching quality monitoring model so as to analyze the relationship between the online data and the teaching quality. According to the invention, teaching data are collected and analyzed in real time, a teaching quality monitoring model is constructed, teaching problems are accurately identified, timely feedback is provided to guide teaching improvement, a monitoring scheme is dynamically adjusted to adapt to teaching changes, and teaching quality is guaranteed and continuously improved.

Description

Technical Field

[0001] The invention relates to the technical field of data processing, and in particular to a method for monitoring the teaching quality of an environmental art design course based on online data. Background Art

[0002] Traditional teaching quality monitoring methods are carried out after the course ends or at a specific time point, and cannot reflect the dynamic changes in the teaching process in real time.

[0003] For example, teachers evaluate students’ learning outcomes through test scores at the end of the semester, but are unable to promptly understand students’ confusion or progress in the learning process, and are therefore unable to adjust teaching strategies in a timely manner.

[0004] In addition, traditional methods focus on the evaluation of overall teaching effects, while ignoring the differences between individual students. Each student has different learning abilities, interests, and design styles, but traditional methods make it difficult to conduct detailed analysis and evaluation of these differences. Teaching quality monitoring relies heavily on teachers’ subjective judgment and experience, and lacks objective and unified standards. This can lead to inconsistency and unfairness in evaluation results.

[0005] For example, in different environmental art design courses, due to different teachers' aesthetic concepts and teaching experience, the evaluation of the same assignment may vary greatly. This subjectivity may affect students' accurate understanding of their own design ability and may also affect the objective evaluation of teaching quality. Summary of the invention

[0006] The technical problem to be solved by the present invention is to provide an environmental art design course teaching quality monitoring method based on online data to achieve real-time monitoring and evaluation of teaching quality.

[0007] In order to solve the above technical problems, the technical solution of the present invention is as follows: In the first aspect, a method for monitoring the teaching quality of an environmental art design course based on online data, the method comprising: Through the online teaching platform, the teaching data of the environmental art design course is collected in real time, including student participation, homework submission, online test scores, classroom interaction records and teacher teaching feedback; Based on the teaching data, a teaching quality monitoring model is constructed through cluster analysis and association rule analysis to analyze the relationship between online data and teaching quality; Through the teaching quality monitoring model, online data is analyzed in real time to identify the group characteristics of different learning performances and teaching styles, and to extract hidden associations in online data to generate data analysis reports; Based on the data analysis report, provide feedback to teachers and students to guide teaching adjustments and optimizations, and monitor the teaching quality of the adjusted environmental art design course in real time. When the teaching quality indicators are monitored to be lower than the preset threshold, the early warning signal is automatically triggered; According to the early warning signals and feedback on teaching effects, the scope, frequency of data collection and parameters of the teaching quality monitoring model are dynamically adjusted.

[0008] Furthermore, the student engagement includes the number of logins, online time, and number of interactions; Assignment submission status includes submission time, completion rate and grade; classroom interaction records include questions, answers and discussions; teacher teaching feedback includes student evaluations and teaching logs.

[0009] Furthermore, based on the teaching data, a teaching quality monitoring model is constructed through cluster analysis and association rule analysis to analyze the relationship between online data and teaching quality, including: Use frequency analysis to extract key features that reflect teaching quality from teaching data, and perform cluster analysis on the key features. Group students or teachers according to learning performance and teaching style dimensions, identify differences between different groups, and obtain cluster analysis results. Based on the cluster analysis results, explore the hidden relationships in the teaching data, including the relationship between high participation and high scores, and the relationship between specific teaching styles and student satisfaction, to obtain association rule analysis results; The group characteristics of cluster analysis results and the hidden relationships of association rule analysis results are used as input data of the teaching quality monitoring model. Through discriminant analysis, the teaching quality monitoring model is constructed to analyze the relationship between online data and teaching quality.

[0010] Furthermore, the teaching quality monitoring model can be used to analyze online data in real time, identify group characteristics of different learning performances and teaching styles, extract hidden associations in online data, and generate data analysis reports, including: A group of individuals that meet the constraints are randomly generated as the initial population. Each individual represents a combination of teaching quality and learning performance. According to the goal of the teaching quality monitoring model, a fitness function is defined to evaluate the pros and cons of each individual to obtain the evaluation results. According to the evaluation results, the final individual is determined as the teacher from the current population according to the individual's fitness value, and the difference between the remaining individuals and the teacher on each decision variable is calculated; The decision variables of the remaining individuals are updated according to the teaching factors and random numbers to obtain the population after the iteration of the teacher stage; In the population after the iteration of the teacher phase, two individuals are randomly selected to compare their fitness values, and the decision variables are updated to obtain the population after the iteration of the learning phase; Analyze the population after the iteration of the learning phase, extract the group characteristics of different learning performances and teaching styles, and identify the hidden associations in the online data to obtain the analysis results of group characteristics and hidden associations; Based on the analysis results of group characteristics and hidden associations, a data analysis report is generated, including the evaluation of the teaching quality monitoring model, the description of the characteristics of different groups, the elaboration of hidden associations and suggestions for improving teaching quality.

[0011] Furthermore, the calculation formula of the fitness function is: ; in, express; , , , represents the weight coefficient; Indicates Secondary homework; Indicates The weight of the task; Indicates the total number of jobs; Indicates the index of the job; Indicates the number of student logins; Indicates the maximum number of logins in the class; Indicates the online time of students; Indicates the maximum online time in the class; Indicates the number of student interactions; Indicates the maximum number of interactions in the class; Indicates the number of times students asked questions; Indicates the maximum number of questions asked in the class; Indicates the number of times students answered; Indicates the maximum number of answers in the class; Indicates the number of times students participated in the discussion; Indicates the maximum number of discussions in the class; Represents a penalty term.

[0012] Furthermore, the calculation formula for the difference between the remaining individuals and the teacher on each decision variable is: ; in, Indicates The individual in The comprehensive difference between the teacher and the decision variables; Indicates The weights of the decision variables; Indicates The individual in The value of the decision variable; Indicates that the teacher The value of the decision variable; , Represents an index variable; An index representing an individual; Indicates The weights of the decision variables; Indicates that it is The individual in The value of the decision variable; Indicates that the teacher The value of the decision variable; Indicates a time; represents a constant; Indicates The individual in The random noise on the decision variables.

[0013] Furthermore, according to the early warning signals and feedback on teaching effects, the scope and frequency of data collection and the parameters of the teaching quality monitoring model are dynamically adjusted, including: Determine the state space and action space in reinforcement learning, and set the reward function, give positive rewards based on teaching effect feedback, and give negative rewards based on early warning signals; According to the current state, select an action from the action space and execute it. After executing the action, observe the changes in teaching effects and warning signals, and calculate and distribute corresponding rewards or penalties according to the reward function. Using the calculated rewards or penalties, the decision-making plan is updated, and the process of collection, selection, execution, observation and update is repeated until the preset number of iterations is reached to determine the final teaching quality monitoring adjustment plan; According to the final teaching quality monitoring adjustment plan, dynamically adjust the scope, frequency of data collection and parameters of the teaching quality monitoring model.

[0014] Furthermore, the calculation formula of the reward function is: ; in, represents the reward function; The weight coefficient representing the improvement of effective teaching effect; represents the number of tests after the teaching action is performed; , Represents an index variable; represents the number of tests before the teaching action is performed; Indicates the first Test scores; Indicates the first The weight of the test; Indicates the first Test scores; Indicates the first The weight of the test; Indicates the adjusted early warning signal weight coefficient; Indicates the number of types of warning signals; Indicates The weight of the early warning signal; Indicates The amount of change in the early warning signal.

[0015] Secondly, the environmental art design course teaching quality monitoring system based on online data includes: A data collection module is used to collect teaching data of the environmental art design course in real time through the online teaching platform. The teaching data includes student participation, homework submission, online test scores, classroom interaction records and teacher teaching feedback; The data analysis and processing module is used to process and analyze the teaching data through cluster analysis and association rule analysis to build a teaching quality monitoring model and analyze the relationship between online data and teaching quality to obtain analysis results; The monitoring report generation module is used to identify the group characteristics of different learning performances and teaching styles based on the analysis results of the data analysis and processing module, extract hidden associations in online data, and generate data analysis reports; Feedback and early warning module, which is used to provide teachers and students with suggestions for teaching adjustment and optimization based on data analysis reports, and monitor the teaching quality of the adjusted environmental art design course in real time. When the teaching quality index is lower than the preset threshold, the early warning signal is automatically triggered; The adjustment module is used to dynamically adjust the data collection range and frequency of the data collection module and the parameters of the teaching quality monitoring model in the data analysis and processing module according to the early warning signals and the feedback of the teaching effect.

[0016] In a third aspect, a computer-readable storage medium stores a program, and when the program is executed by a processor, the method described above is implemented.

[0017] The above solution of the present invention includes at least the following beneficial effects: Through the online teaching platform, real-time collection of teaching data, including student participation, homework submission, online test scores, etc., makes teaching quality monitoring more timely and accurate. Teachers can understand students' learning dynamics and classroom performance at any time, adjust teaching strategies in time, and ensure teaching quality. Using cluster analysis and association rule analysis techniques to build a teaching quality monitoring model, we can deeply explore the hidden associations and group characteristics in online data. This not only helps teachers to fully understand students' learning performance and design style, but also provides strong support for personalized teaching to meet students' differentiated needs.

[0018] The teaching quality monitoring model is based on a large amount of objective data for analysis and evaluation, reducing the impact of human subjective factors. The data analysis report is based on objective data, providing accurate feedback and suggestions for teachers and students, enhancing the objectivity and fairness of teaching quality monitoring. When the teaching quality indicators are monitored to be lower than the preset threshold, the system will automatically trigger an early warning signal to promptly remind teachers and administrators to take measures to intervene. At the same time, based on the early warning signal and feedback on teaching effects, the system can dynamically adjust the scope, frequency of data collection and parameters of the teaching quality monitoring model to achieve intelligent management and optimization.

[0019] By analyzing online data in real time, generating data analysis reports, providing feedback and suggestions, and intelligent early warning and adjustment, this method forms a closed-loop teaching quality monitoring system. This helps teachers continuously optimize teaching strategies and methods, promote the continuous improvement of teaching quality and the all-round development of students. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 It is a flow chart of a method for monitoring the teaching quality of an environmental art design course based on online data provided by an embodiment of the present invention.

[0021] Figure 2 It is a schematic diagram of an environmental art design course teaching quality monitoring system based on online data provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0022] The exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

[0023] like Figure 1 As shown, an embodiment of the present invention proposes a method for monitoring the teaching quality of an environmental art design course based on online data, and the method comprises the following steps: Step 11: Collect teaching data of the environmental art design course in real time through the online teaching platform, including student participation, homework submission, online test scores, classroom interaction records and teacher teaching feedback; Step 12, based on the teaching data, through cluster analysis and association rule analysis, a teaching quality monitoring model is constructed to analyze the relationship between online data and teaching quality; Step 13: Use the teaching quality monitoring model to analyze the online data in real time, identify the group characteristics of different learning performances and teaching styles, extract hidden associations in the online data, and generate a data analysis report; Step 14, based on the data analysis report, provide feedback to teachers and students to guide teaching adjustment and optimization, and monitor the teaching quality of the adjusted environmental art design course in real time. When the teaching quality indicator is monitored to be lower than the preset threshold, an early warning signal is automatically triggered; Step 15, dynamically adjust the scope and frequency of data collection and the parameters of the teaching quality monitoring model according to the early warning signals and the feedback of teaching effect.

[0024] In the embodiment of the present invention, the teaching data is collected in real time through the online teaching platform, ensuring the timeliness and accuracy of the data, which enables teachers and administrators to understand students' learning status, class participation and homework completion in a timely manner.

[0025] Step 12, using cluster analysis and association rule analysis techniques to build a teaching quality monitoring model, can deeply mine the potential information and association relationships in teaching data. This helps to reveal the inherent connection between online data and teaching quality, and provides a scientific basis for teaching quality evaluation.

[0026] Step 13: Real-time analysis of online data through the teaching quality monitoring model can quickly identify the group characteristics of different learning performances and teaching styles. At the same time, it extracts hidden associations in online data and generates detailed data analysis reports to provide teachers and students with intuitive and comprehensive teaching quality feedback.

[0027] Step 14: Based on the data analysis report, provide targeted feedback and suggestions to teachers and students to guide them in adjusting and optimizing teaching. At the same time, monitor the adjusted teaching quality in real time. When the teaching quality indicator is lower than the preset threshold, the early warning signal is automatically triggered to ensure timely intervention measures to prevent the decline of teaching quality.

[0028] Step 15, dynamically adjust the scope, frequency and parameters of the teaching quality monitoring model of data collection according to the early warning signal and the feedback of teaching effect. This enables the teaching quality monitoring system to adapt to the ever-changing teaching environment and needs and maintain its accuracy and effectiveness. At the same time, through continuous adjustment and optimization, the performance and reliability of the teaching quality monitoring system can be further improved.

[0029] In a preferred embodiment of the present invention, the above step 11 collects teaching data of the environmental art design course in real time through the online teaching platform, including student participation, homework submission, online test scores, classroom interaction records and teacher teaching feedback; the student participation includes the number of logins, online time and number of interactions; homework submission includes submission time, completion and score; classroom interaction records include questions, answers and discussions; teacher teaching feedback includes student evaluations and teaching logs, which may include: In an embodiment of the present invention, the types of teaching data that need to be collected are clearly defined, including student participation, homework submission, online test scores, classroom interaction records, and teacher teaching feedback. According to the API interface provided by the online teaching platform, develop or configure corresponding data collection tools, such as data capture scripts, data import modules, etc. The number of logins of each student is recorded through the user login log of the online teaching platform. Set a timed task to regularly obtain the login log from the platform and parse the student's login information. Use the online time statistics function provided by the platform to obtain the student's online time in the classroom. If the platform does not provide a direct statistical function, the online time can be calculated by recording the student's login and exit time. Monitor the student's interactive behavior in the classroom, such as clicking, speaking, voting, etc., and record the number of interactions. Obtain the student's interactive data through the platform's event log or interactive recording interface. When the student submits the homework, the platform automatically records the submission time. Ensure that the platform's time synchronization is accurate to ensure the accuracy of the submission time. The system automatically evaluates the completion of the homework, such as whether it is completed on time, whether it is fully submitted, etc. The completion information is recorded in the homework data in the form of a label or score. After the teacher corrects the homework, the score is entered into the platform. Ensure that grades are entered in a timely and accurate manner and are associated with homework data. Teachers create online tests on the platform and set questions, answers, and scoring criteria. After students complete the test, the platform automatically calculates the grades.

[0030] When students ask questions in class, the platform records the content, time and information of the questioner. When teachers or students answer questions, the platform records the content, time and information of the answerer. Monitor discussion behaviors in class, record discussion topics, participants and discussion content. Through text analysis, conduct preliminary processing of discussion content. After the course, the platform sends evaluation questionnaires to students to collect their evaluation of teachers.

[0031] Teachers record teaching logs during the course, including teaching content, teaching methods, student reactions, etc. Provide online editing and submission functions for teaching logs, so that teachers can record at any time. Clean the collected data to remove duplicate, invalid or erroneous data. Integrate different types of data to form a complete teaching data set. Store the teaching data set in a secure database to ensure data accessibility and security.

[0032] In a preferred embodiment of the present invention, the above step 12, based on the teaching data, constructs a teaching quality monitoring model through cluster analysis and association rule analysis to analyze the relationship between online data and teaching quality, which may include: Step 121, extracting key features reflecting teaching quality from teaching data using frequency analysis, and performing cluster analysis on the key features, grouping students or teachers according to learning performance and teaching style dimensions, identifying differences between different groups, and obtaining cluster analysis results; Step 122, based on the cluster analysis results, exploring hidden relationships in the teaching data, including the relationship between high participation and high scores, and the relationship between a specific teaching style and student satisfaction, to obtain association rule analysis results; Step 123, using the group characteristics of the cluster analysis results and the hidden relationship of the association rule analysis results as input data of the teaching quality monitoring model, and constructing the teaching quality monitoring model through discriminant analysis to analyze the relationship between online data and teaching quality.

[0033] In an embodiment of the present invention, all records in the teaching data set are traversed to check whether each feature has missing values. For features with fewer missing values, the mean, median or mode of the feature is selected for filling. For features with more missing values, or when the missing values ​​have a greater impact on the analysis results, consider deleting the feature or deleting the records containing missing values. Use statistical methods to identify outliers in the data. For outliers that obviously deviate from the normal range, choose to delete or replace them with reasonable values ​​according to the specific situation. Pay attention to retaining outliers that may have practical significance, such as some extreme learning performance or special teaching styles. Check whether there are exactly the same records or highly similar records in the data set. According to the distribution of the data and the analysis requirements, select a standardization method (such as Z score standardization) to convert the data into a distribution with a mean of 0 and a standard deviation of 1. Apply the selected standardization method to each feature to ensure that each feature has the same dimension. Based on the standardized data, calculate the covariance matrix between the features. The covariance matrix reflects the correlation and degree of variation between the features. Perform eigenvalue decomposition on the covariance matrix to solve the eigenvalues ​​and eigenvectors. The eigenvalue indicates the degree of variation in the direction of the corresponding eigenvector. The larger the eigenvalue, the greater the variation in that direction. According to the size of the eigenvalue, the eigenvectors corresponding to the first few larger eigenvalues ​​are selected as the principal components. The principal component is a linear combination feature that can retain the data variation to the greatest extent. Analyze the weights of each original feature in the principal component and identify the key features that have a greater impact on teaching quality. Key features include the number of students' logins, online time, homework completion, test scores, as well as the teacher's teaching style and student evaluation.

[0034] Choose the appropriate number of clusters based on the characteristics of the data and analysis requirements. Randomly select k data points as the initial cluster centers. Assign each data point to the nearest cluster center based on a distance metric (such as Euclidean distance). Calculate the new center of each cluster, that is, the mean of all data points in the cluster, until the cluster center no longer changes, analyze the clustering results, and divide students or teachers into different groups based on dimensions such as their learning performance and teaching style. Each group represents a collection of students or teachers with similar characteristics. Calculate the similarity within the group (such as the within-group mean square error) and the difference between the groups (such as the between-group mean square error ratio). The higher the similarity and the greater the difference, the better the clustering effect.

[0035] Step 122, using the association rule mining algorithm, mine frequent item sets that meet the support threshold. Analyze the composition and distribution of frequent item sets to understand the main patterns and features in the data. Generate association rules that meet the confidence threshold based on frequent item sets. Sort and filter the association rules to mine the most valuable and meaningful rules. Focus on the association between high participation and high scores, and the relationship between a specific teaching style and student satisfaction. By setting specific antecedents and consequents, mine association rules related to these relationships. Explain the mined association rules to clarify the logical relationship between the antecedents and consequents. Analyze the item set composition and confidence values ​​in the rules to understand the reliability and strength of the rules. In-depth exploration of the reasons behind the association rules and possible influencing factors. Combined with teaching practice and expert knowledge, analyze why high participation is associated with high scores, why a specific teaching style affects student satisfaction, etc. Evaluate the possible impact of association rules on teaching practice and decision-making. Analyze the application value of association rules, such as whether they can provide guidance for teaching improvement, whether they can provide a basis for teaching evaluation, etc. The mined association rules are verified using actual teaching data. The accuracy and reliability of the rules are evaluated by comparing the occurrence of the antecedent and consequent items in the rules in the actual data.

[0036] Invite experts in the teaching field to review and verify the mined association rules. According to the opinions and suggestions of experts, adjust and optimize the rules to improve their accuracy and reliability. Analyze the verification results and evaluate the application effect of association rules in actual teaching. Based on the verification results, summarize and reflect on the association rule mining process.

[0037] Step 123, integrate the group characteristics of the cluster analysis results and the hidden relationships of the association rule analysis results to form a comprehensive data set. Ensure that each record in the data set contains group characteristic information and hidden relationship information. According to the analysis objectives and data characteristics, select logistic regression as the discriminant analysis method. Select features that have an impact on teaching quality from the integrated data set as model input. Feature selection should be based on the group characteristics of the cluster analysis results and the hidden relationships of the association rule analysis results to ensure that the features are representative and relevant. Use the selected features to train the logistic regression model. Optimize model performance by adjusting model parameters (such as regularization coefficient, number of iterations, etc.). Use the cross-validation method to verify the model and evaluate the generalization ability of the model. According to the cross-validation results, adjust and optimize the model to improve the accuracy and stability of the model.

[0038] Suppose there is a set of teaching data, including the number of student logins, online time, homework completion, test scores, and the teacher's teaching style (such as clear explanation, frequent interaction, etc.) and student evaluation. Through statistical methods, it is found that the number of logins and online time are positively correlated with students' academic performance, and homework completion is also an important factor affecting grades. At the same time, the clarity of teachers' explanations and the frequency of interaction are closely related to student satisfaction. According to the number of logins, online time and homework completion, students are divided into three groups: "high participation-high scores", "low participation-low scores" and "medium participation-medium scores". According to the clarity of teachers' explanations and the frequency of interaction, teachers are divided into three groups: "clear explanation-frequent interaction", "fuzzy explanation-few interactions" and "average explanation-average interaction". The association rule of "students with high participation (many logins and long online time) often get high scores" is mined. The hidden relationship of "teachers with clear explanations and frequent interactions are more likely to obtain student satisfaction" is found. The group characteristics and association rules are used as input data, and the teaching quality monitoring model is constructed using the logistic regression method. The model can predict students' academic performance and satisfaction, and provide decision support for teaching adjustment and optimization.

[0039] Through cluster analysis, we can accurately identify groups of students with different learning performances, providing strong support for personalized teaching. Association rule mining can reveal hidden relationships in teaching data, helping teachers and administrators understand the inherent laws of the teaching process. By building a teaching quality monitoring model, we can analyze the relationship between online data and teaching quality in real time, providing a scientific basis for teaching decisions. The application of the teaching quality monitoring model can help teachers adjust teaching strategies and methods in a timely manner, improve teaching quality and students' learning outcomes. Through the analysis results of the teaching quality monitoring model, administrators can understand the problems and deficiencies in the teaching process, providing strong support for teaching management and decision-making.

[0040] In a preferred embodiment of the present invention, the above step 13, through the teaching quality monitoring model, performs real-time analysis on the online data, identifies group characteristics of different learning performances and teaching styles, extracts hidden associations in the online data, and generates a data analysis report, which may include: Step 131, randomly generate a group of individuals that meet the constraints as the initial population, each individual represents a combination of teaching quality and learning performance, and define a fitness function according to the goal of the teaching quality monitoring model to evaluate the quality of each individual to obtain an evaluation result; Step 132, according to the evaluation results, the final individual is determined as the teacher from the current population according to the individual's fitness value, and the difference between the remaining individuals and the teacher on each decision variable is calculated; the calculation formula for the difference between the remaining individuals and the teacher on each decision variable is: ; in, Indicates The individual in The comprehensive difference between the teacher and the decision variables; Indicates The weights of the decision variables; Indicates The individual in The value of the decision variable; Indicates that the teacher The value of the decision variable; , Represents an index variable; An index representing an individual; Indicates The weights of the decision variables; Indicates that it is The individual in The value of the decision variable; Indicates that the teacher The value of the decision variable; Indicates a time; represents a constant; Indicates The individual in Random noise on the decision variables; Step 133, updating the decision variables of the remaining individuals according to the teaching factor and the random number to obtain the population after the teacher phase iteration; Step 134, randomly selecting two individuals from the population after the iteration of the teacher phase, comparing the fitness values, and updating the decision variables to obtain the population after the iteration of the learning phase; Step 135, analyzing the population after the iteration of the learning phase, extracting the group characteristics of different learning performances and teaching styles, and identifying the hidden associations in the online data, to obtain the analysis results of the group characteristics and the hidden associations; Step 136, based on the analysis results of group characteristics and hidden associations, a data analysis report is generated, including an evaluation of the teaching quality monitoring model, a description of different group characteristics, an explanation of hidden associations, and suggestions for improving teaching quality.

[0041] In an embodiment of the present invention, a group of individuals are randomly generated according to preset constraints (such as the scoring range of teaching quality, the grading of learning performance, etc.). Each individual is composed of a set of decision variable values, which represent different combinations of teaching quality and learning performance. In order to ensure the diversity of the population, it is ensured that the generated individuals are evenly distributed in the decision variable space. According to the objectives of the teaching quality monitoring model (such as improving students' grades, enhancing learning interest, etc.), a fitness function is defined. This function accepts an individual as input and calculates its fitness value based on the individual's decision variable values. The fitness value is a scalar that indicates the degree of quality of the individual. The higher the value, the more the individual meets the objectives of the model. When defining the fitness function, the weights of different decision variables are considered to reflect their different effects on the fitness value.

[0042] Step 132, based on the fitness values ​​of individuals in the current population, select a teacher based on the individual fitness values. This teacher individual represents the final teaching quality and learning performance combination in the current population. For each remaining individual, calculate its comprehensive difference with the teacher on each decision variable . Indicates The weight of a decision variable reflects the importance of the decision variable in the comprehensive difference calculation. Indicates The individual in The value of a decision variable. Indicates that the teacher The value of a decision variable. , is the index variable used to traverse all decision variables. is the index of an individual. Represents a time that is used to adjust the time effect of the difference between different decision variables. is a constant used to adjust the effect of the direct difference term. Indicates The individual in The random noise on the decision variables is used to simulate the uncertainty factors in actual teaching. By calculating this comprehensive difference, the difference between each individual and the teacher is evaluated.

[0043] Step 133, according to teaching experience, set a teaching factor to control the update speed and direction of individual decision variables. At the same time, generate a random number to introduce randomness to prevent the algorithm from falling into local optimality. For each remaining individual, update its decision variable value according to certain rules based on its current decision variable value, teaching factor and random number. The update rule is linear, ensuring that the updated individual still meets the constraints to ensure the validity of the solution.

[0044] Step 134: randomly select two individuals from the population after the teacher phase iteration for comparison. According to the comparison result, update the decision variables of the two individuals according to certain rules. The update rule is survival of the fittest.

[0045] Step 135, using data mining techniques such as cluster analysis and association rule mining, analyze the population after the iteration of the learning phase. Cluster analysis can divide individuals into different groups and identify individuals with similar characteristics. According to the values ​​of decision variables and fitness values ​​and other information, individuals are clustered into different learning performance and teaching style groups. Association rule mining can discover hidden associations and rules between different decision variables. Analyze the correlation between decision variables and extract association rules that affect teaching quality.

[0046] Based on the results of cluster analysis, group characteristics of different learning performance and teaching styles are extracted. These characteristics can include common features in teaching methods, student participation, homework difficulty, etc. Based on the results of association rule mining, hidden associations in online data are identified. These associations can reveal the intrinsic connection between specific teaching strategies and improved student performance and student satisfaction.

[0047] Step 136, based on the previous analysis results, evaluate the performance of the teaching quality monitoring model. Evaluation indicators may include the accuracy, stability, and interpretability of the model. Describe the extracted characteristics of different groups in detail, including their composition, characteristics, influencing factors, etc. Use charts, statistical data, etc. to intuitively display group characteristics for easy user understanding and analysis. Explain the hidden associations in the identified online data, including the rules, strength, and credibility of the associations. Analyze the potential impact of hidden associations on teaching quality, and put forward corresponding explanations and hypotheses. Based on the analysis results, put forward suggestions for improving teaching quality. Improvement suggestions may include specific measures such as adjusting teaching strategies, improving teaching styles, and optimizing homework design. Ensure that the improvement suggestions are operational and targeted, and can provide effective guidance for actual teaching.

[0048] Suppose there is a teaching quality monitoring model, the goal is to maximize students' test scores and satisfaction. The initial population randomly generates 100 individuals, each of which represents a combination of teaching quality and learning performance (such as teaching methods, student participation, homework difficulty, etc.). Define the fitness function. Randomly generate the initial population, such as individual 1 (teaching method A, high student participation, moderate homework difficulty), individual 2 (teaching method B, low student participation, high homework difficulty), etc. Select the individual with the highest fitness as the teacher (assuming it is individual 1). Calculate the comprehensive difference between the remaining individuals and the teacher in each decision variable, such as the difference between individual 2 and the teacher in teaching method is a certain value, and the difference between individual 2 and the teacher in student participation is another value. According to the teaching factor and random number, update the decision variables of the remaining individuals, such as individual 2's teaching method is adjusted to be closer to individual 1's teaching method, and student participation is improved. In the population after the iteration of the learning phase, randomly select two individuals to compare the fitness values, such as individual 2 and individual 3. According to the comparison results, update their decision variables, such as individual 2's homework difficulty is adjusted to be more moderate. Analyze the population after the iteration of the learning phase, extract the common characteristics of the high-scoring groups (such as teaching method A, high student engagement), and identify the relationship between specific teaching strategies and improved student performance. Generate a data analysis report to evaluate the performance of the model, describe the characteristics of the high-scoring groups, explain the relationship between teaching strategies and student performance, and put forward suggestions for improving teaching quality (such as promoting teaching method A and improving student engagement).

[0049] By analyzing online data in real time and identifying group characteristics of different learning performances and teaching styles, we can understand the current teaching situation more accurately and provide strong support for teaching improvement. Through association rule mining, we can identify hidden associations in online data, such as the association between specific teaching strategies and student performance improvement, which helps to discover teaching rules and improve teaching effectiveness. By continuously iterating and optimizing model parameters, we can improve the accuracy and stability of the model, making the teaching quality monitoring model more reliable and effective. The generated data analysis report can provide strong support for teaching decisions, help teachers and managers understand the current teaching situation, find problems, put forward improvement suggestions, and promote the continuous improvement of teaching quality.

[0050] In a preferred embodiment of the present invention, the calculation formula of the fitness function is: ; in, Represents an individual The fitness value of , , , represents the weight coefficient; Indicates Secondary homework; Indicates The weight of the task; Indicates the total number of jobs; Indicates the index of the job; Indicates the number of student logins; Indicates the maximum number of logins in the class; Indicates the online time of students; Indicates the maximum online time in the class; Indicates the number of student interactions; Indicates the maximum number of interactions in the class; Indicates the number of times students asked questions; Indicates the maximum number of questions asked in the class; Indicates the number of times students answered; Indicates the maximum number of answers in the class; Indicates the number of times students participated in the discussion; Indicates the maximum number of discussions in the class; Represents a penalty term.

[0051] In the embodiment of the present invention, the job completion status part: It is the weight coefficient of the job completion.

[0052] Calculate the sum of all assignment weighted scores , For the The score of the assignment, For the Calculate the sum of all job weights Calculate the weighted average assignment score The contribution of job completion to the fitness value is .

[0053] Student online participation section: is the weight coefficient of students’ participation in online activities. , , They represent the number of students’ logins, online time, and number of interactions respectively. , , Respectively represent the maximum number of logins, online time, and number of interactions in the class. Calculate the relative value of students' online activity participation , reflecting the student’s level of activity in the class. The contribution of student online activity participation to the fitness value is .

[0054] Student question and answer activity section: It is the weight coefficient of students’ questioning and answering activity. , , They represent the number of times students asked questions, answered questions, and participated in discussions. , , Respectively represent the maximum number of questions, answers, and discussions in the class. Calculate the relative value of students' questioning and answering activity , reflecting the students’ knowledge exchange and participation in the class. The contribution of students’ questioning and answering activity to the fitness value is .

[0055] Penalty section: is the weight coefficient of the penalty term. It is a penalty item used to deduct points for behaviors that do not meet teaching requirements or regulations. is the negative impact of the penalty term on the fitness value.

[0056] Add the homework completion part, the student online activity participation part, the student question and answer activity part and the penalty part to get the individual fitness value The higher the fitness value, the better the individual's teaching quality and learning performance.

[0057] The fitness function comprehensively considers multiple aspects such as homework completion, student participation in online activities, and student questioning and answering activity, and can comprehensively and objectively evaluate individual teaching quality and learning performance.

[0058] By setting weight coefficients and penalty items, the fitness function can motivate individuals to actively complete homework, participate in online activities and knowledge exchange, and improve teaching effectiveness and learning enthusiasm. The fitness function provides a clear guiding direction for the optimization of teaching quality. By analyzing the changes in fitness values, the key factors affecting teaching quality can be found, providing strong support for teaching improvement.

[0059] In a preferred embodiment of the present invention, the above step 14 provides feedback to teachers and students based on the data analysis report, guides teaching adjustment and optimization, and monitors the teaching quality of the adjusted environmental art design course in real time. When the teaching quality index is monitored to be lower than the preset threshold, an early warning signal is automatically triggered, which may include: In an embodiment of the present invention, the teaching data of the environmental art design course (including student participation, homework submission, online test scores, classroom interaction records and teacher teaching feedback) collected in real time from the online teaching platform are integrated to ensure the integrity and consistency of the data. The integrated data is analyzed in depth using data mining techniques such as cluster analysis and association rule analysis. Cluster analysis is used to identify group characteristics of different learning performances and teaching styles; association rule analysis is used to extract hidden associations in online data, such as the correlation between student participation and homework submission. According to the data analysis results, a data analysis report is written. The report should include an analysis of key indicators, a description of group characteristics, the revelation of hidden associations, and suggestions for teaching improvements. The key findings and suggestions in the data analysis report are presented to teachers in an intuitive and easy-to-understand manner, such as through charts, tables, etc. At the same time, regular teacher meetings or seminars are organized to discuss the content of the report and promote communication and learning among teachers. For students, a simplified summary of the data analysis report can be published through the online teaching platform, or targeted learning suggestions can be provided using a personalized learning system. In addition, students' feedback on teaching quality can also be collected through classroom discussions, questionnaires, etc.

[0060] According to the teaching objectives and requirements of the environmental art design course, a series of teaching quality indicators are set, such as student participation indicators, homework submission rate indicators, online test score indicators, etc. Use the real-time data collection function of the online teaching platform to continuously monitor the changes of these indicators. You can set up a scheduled task or event trigger mechanism to ensure real-time update and monitoring of data. Set reasonable preset thresholds for each teaching quality indicator. These thresholds should be dynamically adjusted based on historical data, teaching objectives, and actual conditions.

[0061] When a certain teaching quality indicator is detected to be lower than its preset threshold, the system automatically makes a judgment and triggers an early warning signal. The early warning signal can be sent to relevant personnel through various means such as email, text message, and online teaching platform notification. After the early warning signal is triggered, the emergency response mechanism should be activated immediately. Organize relevant personnel to conduct in-depth analysis of the cause of the early warning, formulate targeted improvement measures, and track the implementation effect to ensure that the teaching quality is improved in a timely manner. Use data analysis reports, early warning signals, and emergency response results as important bases for teaching adjustments to form a feedback loop. By continuously collecting and analyzing data, providing feedback, monitoring teaching quality, triggering early warning signals, and adjusting teaching strategies, continuous improvement and optimization of teaching quality can be achieved. According to the monitoring results and feedback of teaching quality, the teaching quality monitoring model is regularly updated and optimized. Adjust the scope, frequency, and parameters of data collection to improve the accuracy and reliability of the model.

[0062] In a preferred embodiment of the present invention, the above step 15, dynamically adjusting the scope and frequency of data collection and the parameters of the teaching quality monitoring model according to the early warning signal and the feedback of the teaching effect, may include: Step 151, determining the state space and action space in reinforcement learning, and setting a reward function, giving positive rewards according to teaching effect feedback, and giving negative rewards according to early warning signals; Step 152, according to the current state, select an action from the action space and execute it, and after executing the action, observe the changes in the teaching effect and the warning signal, and calculate and distribute the corresponding reward or punishment according to the reward function; Step 153, using the calculated reward or penalty, update the decision plan, and repeat the process of collecting, selecting, executing, observing and updating until the preset number of iterations is reached to determine the final teaching quality monitoring adjustment plan; Step 154, dynamically adjust the scope and frequency of data collection and the parameters of the teaching quality monitoring model according to the final teaching quality monitoring adjustment plan.

[0063] In an embodiment of the present invention, the state space contains all possible teaching states, such as student participation, homework completion rate, test score distribution, triggering of early warning signals, etc. Each state is represented by a set of feature vectors, which can be extracted from data sources such as teaching management systems and online learning platforms. The action space contains all possible adjustment actions, such as increasing the frequency of data collection, expanding the scope of data collection, adjusting the parameters of the teaching quality monitoring model, etc. Each action has a clear execution method and expected effect, such as "adjusting the data collection frequency from once an hour to once every half an hour." Positive rewards are given based on feedback on teaching effects, such as improved student grades and increased participation. Negative rewards are given based on early warning signals, such as decreased student attendance and decreased homework submission rates. The reward function can be linear or nonlinear to accurately reflect the impact of teaching effects and early warning signals on teaching quality.

[0064] Step 152, select an action from the action space according to the current state. Perform the selected action, such as adjusting the frequency or range of data collection, or modifying the parameters of the teaching quality monitoring model. After performing the action, observe the changes in the teaching effect and the early warning signal. This can be achieved by comparing the data before and after the action is performed. According to the reward function, calculate and distribute the corresponding reward or penalty. The value of the reward or penalty will be used to update the decision plan.

[0065] Step 153, using the calculated reward or penalty, update the decision plan. This can be achieved by adjusting the action selection strategy, modifying the reward function, or optimizing the state space representation. The process of collecting, selecting, executing, observing, and updating is repeated until a preset number of iterations is reached or a certain stop condition is met. After reaching the preset number of iterations or meeting the stop condition, the final teaching quality monitoring adjustment plan is determined.

[0066] Step 154, dynamically adjust the scope and frequency of data collection and the parameters of the teaching quality monitoring model according to the final teaching quality monitoring adjustment plan, so as to ensure that the adjusted plan can more effectively monitor the teaching quality, respond to early warning signals in a timely manner, and improve the teaching effect.

[0067] Suppose there is an online learning platform. In order to evaluate the teaching effect, it is necessary to closely monitor the student participation and homework completion rate. In the initial state, the system collects data once an hour, and the scope of collection is limited to the student's login record and homework submission. The state space of the system is composed of key indicators such as the student's login frequency, homework submission rate, and online time. The action space includes increasing the frequency of data collection and expanding the scope of data collection (for example, including the number of students' interactions and questions). The reward function is designed to give positive rewards when the student's homework submission rate increases, and negative rewards when the student's login frequency decreases.

[0068] Current Status and Analysis: Student login frequency: Moderate, 3-4 times a day on average.

[0069] Assignment submission rate: 75%, some students delayed submission or did not submit.

[0070] Online time: The average online time of students is 2 hours per day. After analyzing the current status, it is decided to take action to improve the teaching effect. According to the current status and the preset strategy, "Increase the frequency of data collection" is selected as the action. The goal is to capture changes in student behavior more timely so as to respond to possible teaching problems more quickly. Adjust the frequency of data collection from once an hour to once every half an hour. After executing the action, continue to monitor the students' homework submission rate and login frequency. After a week of observation, it was found that the homework submission rate increased to 80%, and more students submitted their homework on time. At the same time, the student login frequency remained stable and did not show a significant decrease. According to the reward function, the system calculates the reward after executing the action. Due to the increase in homework submission rate, the system gives a positive reward. Use the reward obtained to update the decision plan. Since the action of increasing the frequency of data collection brings positive teaching effects, the system will be more inclined to choose this action in future decisions. The system adjusts the action selection strategy so that it is more likely to choose to increase the frequency of data collection when facing a similar state next time. Continue to collect new teaching status data, select actions according to the updated decision plan, execute the actions, and observe the changes. After many iterations, the system gradually optimizes the teaching quality monitoring strategy. When the preset number of iterations was reached, the final adjustment plan for teaching quality monitoring was determined. According to the final plan, the frequency of data collection was adjusted to once every half hour, and the scope of data collection was expanded to include the number of students' interactions and questions.

[0071] By increasing the frequency and scope of data collection, the system can more comprehensively evaluate teaching effectiveness and provide better teaching support and services for teachers and students. By dynamically adjusting the scope and frequency of data collection and the parameters of the teaching quality monitoring model, the system can better adapt to changes in the teaching process and respond to early warning signals in a timely manner. Through the reinforcement learning mechanism, the system can continuously optimize decision-making plans to improve teaching effectiveness, such as improving students' homework submission rate, participation and test scores. The automated adjustment process reduces the need for manual intervention and reduces the workload of teachers and administrators, allowing them to focus more on teaching itself. Through continuous iteration and optimization, the system's ability to respond to abnormal situations and early warning signals is enhanced, improving the robustness and reliability of the system.

[0072] The decision-making process based on data collection and analysis is more objective and scientific, and helps to improve the level and efficiency of teaching management.

[0073] In a preferred embodiment of the present invention, the calculation formula of the reward function is: ; in, represents the reward function; The weight coefficient representing the improvement of effective teaching effect; represents the number of tests after the teaching action is performed; , Represents an index variable; represents the number of tests before the teaching action is performed; Indicates the first Test scores; Indicates the first The weight of the test; Indicates the first Test scores; Indicates the first The weight of the test; Indicates the adjusted early warning signal weight coefficient; Indicates the number of types of warning signals; Indicates The weight of the early warning signal; Indicates The amount of change in the early warning signal.

[0074] In the embodiment of the present invention, various parameters in the teaching quality monitoring system are initialized, including the weight coefficient of the effective teaching effect improvement amount , Adjusted warning signal weight coefficient , as well as the initial values ​​of various test scores, weights and warning signals. At the same time, determine the initial frequency and range of data collection. According to the current teaching status and preset strategies, select "Increase the frequency of data collection" as the teaching action. After executing this action, the system will collect relevant data such as student participation and homework completion rate at a new frequency (such as once every half an hour). Before and after executing the teaching action, collect a certain amount of test data (such as students' homework grades, online test scores, etc.) and warning signals (such as a decrease in student login frequency, delayed homework submission, etc.). Suppose the number of tests before executing the teaching action is , the number of tests after execution is For each test, record the score or And the corresponding weights or At the same time, record the number of types of various warning signals , weight And the amount of change . Calculate the reward value after executing the teaching action according to the formula of the reward function The details are as follows: Calculate the weighted average of the test scores after performing the teaching action Calculate the weighted average of the test scores before performing the teaching action Calculate the difference between the two and multiply it by the weight coefficient of the effective teaching effect improvement . Calculate the weighted sum of all warning signal changes .Will Multiply by the adjusted early warning signal weight coefficient , and subtract it from the previous result. Get the reward value . According to the reward value , update the decision-making plan of the teaching quality monitoring system. If is positive, it means that the executed teaching action has a positive impact on the teaching effect, and the system will be more inclined to choose this action in the future. If it is negative, it means that the teaching action may have a negative impact, and the system needs to adjust its strategy and choose other actions that may be more effective. Continue to perform teaching actions, collect data, calculate rewards, and update decision plans. After multiple iterations, the final teaching action combination is determined.

[0075] Through the design of the reward function, it is possible to automatically identify and select teaching actions that have a positive impact on teaching results, thereby improving the overall teaching quality. The reward function takes into account the change in the warning signal, so that the system can respond promptly to possible teaching problems, such as decreased student participation and delayed homework submission. The system can automatically make decisions based on the current teaching status and reward function, reducing the need for manual intervention and improving the efficiency and accuracy of decision-making. Through continuous iterative optimization, it can gradually adapt to different teaching environments and needs and find the most suitable teaching strategy for the current situation. The calculation of the reward function is based on actual teaching data and warning signals, making the decision more scientific, reasonable and reliable.

[0076] like Figure 2 As shown, the embodiment of the present invention also provides an environmental art design course teaching quality monitoring system based on online data, including: A data collection module is used to collect teaching data of the environmental art design course in real time through the online teaching platform. The teaching data includes student participation, homework submission, online test scores, classroom interaction records and teacher teaching feedback; The data analysis and processing module is used to process and analyze the teaching data through cluster analysis and association rule analysis to build a teaching quality monitoring model and analyze the relationship between online data and teaching quality to obtain analysis results; The monitoring report generation module is used to identify the group characteristics of different learning performances and teaching styles based on the analysis results of the data analysis and processing module, extract hidden associations in online data, and generate data analysis reports; Feedback and early warning module, which is used to provide teachers and students with suggestions for teaching adjustment and optimization based on data analysis reports, and monitor the teaching quality of the adjusted environmental art design course in real time. When the teaching quality index is lower than the preset threshold, the early warning signal is automatically triggered; The adjustment module is used to dynamically adjust the data collection range and frequency of the data collection module and the parameters of the teaching quality monitoring model in the data analysis and processing module according to the early warning signals and the feedback of the teaching effect.

[0077] It should be noted that the system is a system corresponding to the above method, and all implementation methods in the above method embodiment are applicable to this embodiment and can achieve the same technical effect.

[0078] The embodiment of the present invention also provides a computer-readable storage medium storing instructions, which, when executed on a computer, enable the computer to execute the method described above. All implementations in the above method embodiment are applicable to this embodiment and can achieve the same technical effect.

[0079] The above is a preferred embodiment of the present invention. It should be pointed out 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 should also be regarded as the scope of protection of the present invention.

Claims

1. A method for monitoring the teaching quality of environmental art design courses based on online data, characterized in that: The method comprises: Through the online teaching platform, the teaching data of the environmental art design course is collected in real time, including student participation, homework submission, online test scores, classroom interaction records and teacher teaching feedback; Based on the teaching data, a teaching quality monitoring model is constructed through cluster analysis and association rule analysis to analyze the relationship between online data and teaching quality; Through the teaching quality monitoring model, online data is analyzed in real time to identify the group characteristics of different learning performances and teaching styles, and to extract hidden associations in online data to generate data analysis reports; Based on the data analysis report, provide feedback to teachers and students to guide teaching adjustments and optimizations, and monitor the teaching quality of the adjusted environmental art design course in real time. When the teaching quality indicators are monitored to be lower than the preset threshold, the early warning signal is automatically triggered; According to the early warning signals and feedback on teaching effects, the scope, frequency of data collection and parameters of the teaching quality monitoring model are dynamically adjusted.

2. The method for monitoring the teaching quality of environmental art design courses based on online data according to claim 1 is characterized in that: The student engagement mentioned includes the number of logins, online time and number of interactions; Assignment submission status includes submission time, completion rate and grade; classroom interaction records include questions, answers and discussions; teacher teaching feedback includes student evaluations and teaching logs.

3. The method for monitoring the teaching quality of environmental art design courses based on online data according to claim 2 is characterized in that: Based on the teaching data, a teaching quality monitoring model is constructed through cluster analysis and association rule analysis to analyze the relationship between online data and teaching quality, including: Use frequency analysis to extract key features that reflect teaching quality from teaching data, and perform cluster analysis on the key features. Group students or teachers according to learning performance and teaching style dimensions, identify differences between different groups, and obtain cluster analysis results. Based on the cluster analysis results, explore the hidden relationships in the teaching data, including the relationship between high participation and high scores, and the relationship between specific teaching styles and student satisfaction, to obtain association rule analysis results; The group characteristics of cluster analysis results and the hidden relationships of association rule analysis results are used as input data of the teaching quality monitoring model. Through discriminant analysis, a teaching quality monitoring model is constructed to analyze the relationship between online data and teaching quality.

4. The method for monitoring the teaching quality of environmental art design courses based on online data according to claim 3 is characterized in that: Through the teaching quality monitoring model, online data is analyzed in real time to identify the group characteristics of different learning performances and teaching styles, and to extract hidden associations in online data to generate data analysis reports, including: A group of individuals that meet the constraints are randomly generated as the initial population. Each individual represents a combination of teaching quality and learning performance. According to the goal of the teaching quality monitoring model, a fitness function is defined to evaluate the pros and cons of each individual to obtain the evaluation results. According to the evaluation results, the final individual is determined as the teacher from the current population according to the individual's fitness value, and the difference between the remaining individuals and the teacher on each decision variable is calculated; The decision variables of the remaining individuals are updated according to the teaching factors and random numbers to obtain the population after the iteration of the teacher stage; In the population after the iteration of the teacher phase, two individuals are randomly selected to compare their fitness values, and the decision variables are updated to obtain the population after the iteration of the learning phase; Analyze the population after the iteration of the learning phase, extract the group characteristics of different learning performances and teaching styles, and identify the hidden associations in the online data to obtain the analysis results of group characteristics and hidden associations; Based on the analysis results of group characteristics and hidden associations, a data analysis report is generated, including the evaluation of the teaching quality monitoring model, the description of the characteristics of different groups, the elaboration of hidden associations and suggestions for improving teaching quality.

5. The method for monitoring the teaching quality of environmental art design courses based on online data according to claim 4 is characterized in that: The calculation formula of the fitness function is: ； in, Represents an individual The fitness value of , , , represents the weight coefficient; Indicates Secondary homework; Indicates The weight of the task; Indicates the total number of jobs; Indicates the index of the job; Indicates the number of student logins; Indicates the maximum number of logins in the class; Indicates the online time of students; Indicates the maximum online time in the class; represents the number of student interactions; Indicates the maximum number of interactions in the class; Indicates the number of times students asked questions; Indicates the maximum number of questions asked in the class; Indicates the number of times students answered; Indicates the maximum number of answers in the class; Indicates the number of times students participated in the discussion; Indicates the maximum number of discussions in the class; Represents a penalty term.

6. The method for monitoring the teaching quality of environmental art design courses based on online data according to claim 5 is characterized in that: The calculation formula for the difference between the remaining individuals and the teacher on each decision variable is: ； in, Indicates The individual in The comprehensive difference between the teacher and the decision variables; Indicates The weights of the decision variables; Indicates The individual in The value of the decision variable; Indicates that the teacher The value of the decision variable; , Represents an index variable; An index representing an individual; Indicates The weights of the decision variables; Indicates that it is The individual in The value of the decision variable; Indicates that the teacher The value of the decision variable; Indicates a time; represents a constant; Indicates The individual in The random noise on the decision variables.

7. The method for monitoring the teaching quality of environmental art design courses based on online data according to claim 6 is characterized in that: According to early warning signals and feedback on teaching effects, dynamically adjust the scope and frequency of data collection and the parameters of the teaching quality monitoring model, including: Determine the state space and action space in reinforcement learning, and set the reward function, give positive rewards based on teaching effect feedback, and give negative rewards based on early warning signals; According to the current state, select an action from the action space and execute it. After executing the action, observe the changes in teaching effects and warning signals, and calculate and distribute corresponding rewards or penalties according to the reward function. Using the calculated rewards or penalties, the decision-making plan is updated, and the process of collection, selection, execution, observation and update is repeated until the preset number of iterations is reached to determine the final teaching quality monitoring adjustment plan; According to the final teaching quality monitoring adjustment plan, dynamically adjust the scope, frequency of data collection and parameters of the teaching quality monitoring model.

8. The method for monitoring the teaching quality of environmental art design courses based on online data according to claim 7 is characterized in that: The reward function is calculated as: ； in, represents the reward function; The weight coefficient representing the improvement of effective teaching effect; represents the number of tests after the teaching action is performed; , Represents an index variable; represents the number of tests before the teaching action is performed; Indicates the first Test scores; Indicates the first The weight of the test; Indicates the first Test scores; Indicates the first The weight of the test; Indicates the adjusted early warning signal weight coefficient; Indicates the number of types of warning signals; Indicates The weight of the early warning signal; Indicates The amount of change in the early warning signal.

9. A system for monitoring the teaching quality of environmental art design courses based on online data, the system implementing the method as claimed in any one of claims 1 to 8, characterized in that: include: A data collection module is used to collect teaching data of the environmental art design course in real time through the online teaching platform. The teaching data includes student participation, homework submission, online test scores, classroom interaction records and teacher teaching feedback; The data analysis and processing module is used to process and analyze the teaching data through cluster analysis and association rule analysis to build a teaching quality monitoring model and analyze the relationship between online data and teaching quality to obtain analysis results; The monitoring report generation module is used to identify the group characteristics of different learning performances and teaching styles based on the analysis results of the data analysis and processing module, extract hidden associations in online data, and generate data analysis reports; Feedback and early warning module, which is used to provide teachers and students with suggestions for teaching adjustment and optimization based on data analysis reports, and monitor the teaching quality of the adjusted environmental art design course in real time. When the teaching quality index is lower than the preset threshold, the early warning signal is automatically triggered; The adjustment module is used to dynamically adjust the data collection range and frequency of the data collection module and the parameters of the teaching quality monitoring model in the data analysis and processing module according to the early warning signals and feedback on the teaching effect.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a program, which implements the method according to any one of claims 1 to 8 when executed by a processor.

Images