School course quality evaluation method and system based on artificial intelligence

Abstract

The invention relates to a school course quality evaluation method based on artificial intelligence, and the method comprises the following steps: S1, determining evaluation dimensions and personalized evaluation indexes: the evaluation dimensions comprise course childbearing guidance, course content structures, academic quality standards and course implementation and effects, and the personalized evaluation indexes comprise subjects and grades; s2, data acquisition: acquiring data according to the evaluation dimension and the personalized evaluation index; s3, data processing: processing and analyzing the collected data according to the evaluation dimension; s4, data evaluation: performing preliminary classification on the data processed and analyzed in the S3 by using a support vector machine, and on the basis of the preliminary classification of the support vector machine, refining a classification result by using a random forest in combination with a personalized evaluation index to obtain a course quality evaluation report of the target school; and S5, result explanation and application suggestion: explaining and explaining the evaluation result, and proposing targeted course improvement suggestions according to the evaluation result.

Description

Technical Field

[0001] The present invention relates to the field of educational technology, and in particular to a school curriculum quality assessment method and system based on artificial intelligence. Background Art

[0002] With the rapid development of society and the continuous advancement of science and technology, people have higher and higher requirements for the quality of education. As an important place to cultivate future pillars of society, the quality of its courses is directly related to students' knowledge mastery, ability cultivation and improvement of comprehensive quality. Therefore, in order to meet the society's demand for high-quality talents, schools must strengthen the evaluation of course quality to ensure the scientific nature of teaching content, the effectiveness of teaching methods and the achievement of teaching goals.

[0003] In the existing technology, different subjects have different teaching characteristics and requirements, and students of different grades have different cognitive levels and learning abilities. Traditional course quality evaluation methods often fail to fully consider subject differences and grade differences. Therefore, it is particularly important to propose a school course quality evaluation method and system based on artificial intelligence. Summary of the invention

[0004] The purpose of the present invention is to solve the shortcomings of the traditional course quality assessment method in the prior art that it is often difficult to fully consider the differences in subjects and grades, and to provide a school course quality assessment method and system based on artificial intelligence.

[0005] In order to achieve the above object, the present invention adopts the following technical solutions: A school curriculum quality assessment method based on artificial intelligence, comprising the following steps: S1: Determine the evaluation dimensions and personalized evaluation indicators: The evaluation dimensions include the curriculum education orientation, curriculum content structure, academic quality standards, and curriculum implementation and effects. The personalized evaluation indicators include disciplines and grades. S2: Data collection: Collect data based on evaluation dimensions and personalized evaluation indicators; S3: Data processing: Process and analyze the collected data according to the evaluation dimensions; S4: Data evaluation: Use support vector machine to preliminarily classify the data processed and analyzed in S3. Based on the preliminary classification by support vector machine, use random forest and combine personalized evaluation indicators to refine the classification results and obtain the course quality evaluation report of the target school. S5: Interpretation of results and application suggestions: Explain and illustrate the evaluation results, including evaluation methods, evaluation indicators, evaluation results, etc., and make targeted course improvement suggestions based on the evaluation results, such as optimizing teaching methods, improving teaching quality, etc.

[0006] The above scheme further comprises: Furthermore, the collected data is processed and analyzed according to the educational orientation of the course. Through content analysis, it is checked whether the core literacy training objectives are clearly put forward in the course, and the specific manifestation of the training objectives in the course is observed. Key information and indicators are extracted, and through student surveys or tests, students' feedback on the course content is collected, their understanding and acceptance are evaluated, and a student feedback report is formed.

[0007] Furthermore, the collected data are processed and analyzed according to the course content structure. By analyzing whether the course content strengthens the connection with the educational goals, in the arrangement of the course content, the internal connection between each knowledge point and the corresponding relationship between each knowledge point and the educational goals are analyzed.

[0008] Furthermore, the collected data is processed and analyzed according to the academic quality standards. By collecting and analyzing the statistical data of students' academic performance and core literacy, an evaluation report on the implementation effect of the academic quality standards is formed. By comparing the academic quality standards and course objectives, their consistency is evaluated. The consistency is expressed as ,in, Indicates consistency, Indicates academic performance, represents the course objective, and n represents the number of evaluation samples.

[0009] Furthermore, the collected data will be processed and analyzed based on the course implementation and effects. Through student satisfaction surveys, students’ feedback and evaluation on the course implementation effects will be collected. By collecting and analyzing statistical data on students’ academic performance and core literacy improvement, the role of course implementation in students’ all-round development will be evaluated.

[0010] Further, in S4, the specific steps of the data evaluation are: S41: Data preparation and processing: Receive the data processed and analyzed in S3, and extract key features and labels from the processed and analyzed data. For example, for language compliance, features such as word accuracy, moderate speech speed, and frequency of unclear semantic expressions can be extracted; for blackboard compliance, features such as formula derivation accuracy, graphic drawing clarity, and blackboard logic can be extracted; S42: Support vector machine preliminary classification: The extracted key features and labels are input into the trained support vector machine model for preliminary classification; S43: Random forest refined classification: Use the preliminary classification results and personalized evaluation indicators as training data to train the random forest model, and input the preliminary classification results into the trained random forest model for refined classification; S44: Derive the evaluation report: Integrate the results of the support vector machine preliminary classification and the random forest refined classification to form a complete course quality evaluation report.

[0011] Further, in S42, the specific steps of the support vector machine preliminary classification are: S421: Input data: input the extracted key features and labels into the support vector machine model; S422: Learning algorithm: The support vector machine model uses SMO to find the best separation hyperplane, the separation hyperplane is in the form of ,in, is the weight vector, is the input feature vector, is a bias term, which makes samples of different categories as correctly classified as possible. The SMO algorithm is an iterative algorithm for solving the dual problem of support vector machines. In each iteration, the SMO algorithm selects two Lagrange multipliers to update to keep the KKT condition satisfied, which is expressed as: Calculate the gradient of L(α): Calculate the gradient of the Lagrangian function L(α) with respect to and The gradient of , the Lagrangian function L(α) is defined as in, is the kernel function, which is used to calculate the similarity between two samples. and The gradient of , Here λ is the inverse of the regularization parameter C, that is, λ = 1 / C; Solve the quadratic programming problem: Simplify the problem into a quadratic programming problem to find the new and values, which will maximize L(α) under the constraints. The quadratic programming problem is expressed as , where the elements of the B matrix are It is related to the kernel function and the label. c is a constant. Solving this quadratic programming problem can get the new and The value of renew and : Updates with the new value found by the solver and , the updated value must satisfy the constraints of SVM, which are expressed as ; Calculate the new value of b: and After that, a support vector is selected (i.e. or samples greater than 0) to calculate the new value of b, , where w is the weight vector, expressed as , select a support vector , so that >0, then Substitute the expression for w into the above formula to get the new value of b; Check KKT conditions, check updated , Do a and b satisfy the KKT condition. If not, continue iterating; if they do, the algorithm converges and finds the best separating hyperplane; S423: Model evaluation: After training is completed, use the test dataset to evaluate the performance of the model; S424: Preliminary classification: Input the new data into the trained support vector machine model to obtain preliminary classification results.

[0012] Further, in S43, the specific steps of the random forest refinement classification are: S431: Feature selection: Select features useful for classification from the preliminary classification results and personalized evaluation indicators, and select the feature with the smallest Gini impurity as the feature useful for classification. The Gini impurity is expressed as ,in, is the number of categories, is the probability of the i-th category; S432: Construct decision tree: For each tree in the random forest, a training set with the same size as the original data set is extracted from the training data set using the Bootstrap method. At each split point, a part of the features is randomly selected for inspection, and the optimal feature is selected to split the node based on the Gini impurity. The above process is repeated until the stopping condition is met; S433: Ensemble decision: build multiple trees, each of which is trained independently. For new samples to be classified, use each decision tree in the random forest to classify them, and decide the final category of the new sample by majority voting; S434: Input preliminary classification results: The preliminary classification results are used as input data and input into the trained random forest model; S435: Obtain refined classification results: The random forest model outputs refined classification results based on the characteristics of the input data.

[0013] A campus resource sharing system used by a school curriculum quality assessment method based on artificial intelligence, comprising: Data collection module: responsible for collecting data related to school curriculum quality assessment; Data processing module: responsible for pre-processing the collected data and processing and analyzing the data related to the school curriculum quality assessment according to the evaluation dimensions; Model evaluation module: responsible for using support vector machines to perform preliminary classification of processed and analyzed data. Based on the preliminary classification of the support vector machine, the random forest is used in combination with personalized evaluation indicators to refine the classification results. Result interpretation and application module: responsible for interpreting and explaining the evaluation results, and making targeted course improvement suggestions based on the evaluation results.

[0014] The present invention has the following beneficial effects: In the present invention, full consideration is given to differences in subjects and grades, and personalized evaluation indicators are formulated. This makes the evaluation results more in line with the actual situation and can more accurately reflect the teaching characteristics and requirements of different subjects and grades. The processed and analyzed data are preliminarily classified using a support vector machine. On the basis of the preliminary classification of the support vector machine, the classification results are refined using a random forest combined with personalized evaluation indicators to obtain a course quality evaluation report for the target school, which can more accurately reflect the actual situation of the course quality, avoid the interference of human factors, and improve the objectivity of the evaluation. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 A method step diagram of a school curriculum quality assessment method and system based on artificial intelligence proposed by the present invention; Figure 2 A diagram showing the steps of the data evaluation method in the present invention. DETAILED DESCRIPTION

[0016] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0017] See also Figure 1-Figure 2 As shown, the present invention is a school curriculum quality assessment method based on artificial intelligence, comprising the following steps: S1: Determine the evaluation dimensions and personalized evaluation indicators: The evaluation dimensions include the curriculum education orientation, curriculum content structure, academic quality standards, and curriculum implementation and effects. The personalized evaluation indicators include disciplines and grades. S2: Data collection: Collect data based on evaluation dimensions and personalized evaluation indicators; S3: Data processing: Process and analyze the collected data according to the evaluation dimensions; S4: Data evaluation: Use support vector machine to preliminarily classify the data processed and analyzed in S3. Based on the preliminary classification by support vector machine, use random forest and combine personalized evaluation indicators to refine the classification results and obtain the course quality evaluation report of the target school. S5: Interpretation of results and application suggestions: Explain and illustrate the evaluation results, including evaluation methods, evaluation indicators, evaluation results, etc., and make targeted course improvement suggestions based on the evaluation results, such as optimizing teaching methods, improving teaching quality, etc.

[0018] In one embodiment, for the above-mentioned course education orientation, the collected data is processed and analyzed according to the course education orientation. Through content analysis, it is checked whether the core literacy training objectives are clearly put forward in the course, and the specific manifestation of the training objectives in the course is observed. Key information and indicators are extracted, and through student surveys or tests, student feedback on the course content is collected, their understanding and acceptance are evaluated, and a student feedback report is formed.

[0019] In one embodiment, for the above-mentioned course content structure, the collected data is processed and analyzed according to the course content structure, and whether the course content enhances the connection with the educational goals is analyzed. In the arrangement of the course content, the internal connection between each knowledge point and the correspondence between each knowledge point and the educational goals are analyzed.

[0020] In one embodiment, for the above-mentioned academic quality standards, the collected data is processed and analyzed according to the academic quality standards, and an evaluation report on the implementation effect of the academic quality standards is formed by collecting and statistically analyzing the statistical data of students' academic performance and core literacy achievement. By comparing the academic quality standards and course objectives, their consistency is evaluated, and the consistency is expressed as ,in, Indicates consistency, Indicates academic performance, represents the course objective, and n represents the number of evaluation samples.

[0021] In one embodiment, for the above-mentioned course implementation and effects, the collected data is processed and analyzed based on the course implementation and effects, and students' feedback and evaluation on the course implementation effects are collected through student satisfaction surveys. By collecting and analyzing statistical data on students' academic performance and core literacy improvement, the role of course implementation in students' all-round development is evaluated.

[0022] In one embodiment, for the above S4, in S4, the specific steps of the data evaluation are: S41: Data preparation and processing: Receive the data processed and analyzed in S3, and extract key features and labels from the processed and analyzed data. For example, for language compliance, features such as word accuracy, moderate speech speed, and frequency of unclear semantic expressions can be extracted; for blackboard compliance, features such as formula derivation accuracy, graphic drawing clarity, and blackboard logic can be extracted; S42: Support vector machine preliminary classification: The extracted key features and labels are input into the trained support vector machine model for preliminary classification; S43: Random forest refined classification: Use the preliminary classification results and personalized evaluation indicators as training data to train the random forest model, and input the preliminary classification results into the trained random forest model for refined classification; S44: Derive the evaluation report: Integrate the results of the support vector machine preliminary classification and the random forest refined classification to form a complete course quality evaluation report.

[0023] In one embodiment, for the above S42, in S42, the specific steps of the support vector machine preliminary classification are: S421: Input data: input the extracted key features and labels into the support vector machine model; S422: Learning algorithm: The support vector machine model uses SMO to find the best separation hyperplane, the separation hyperplane is in the form of ,in, is the weight vector, is the input feature vector, is a bias term, which makes samples of different categories as correctly classified as possible. The SMO algorithm is an iterative algorithm for solving the dual problem of support vector machines. In each iteration, the SMO algorithm selects two Lagrange multipliers to update to keep the KKT condition satisfied, which is expressed as: Calculate the gradient of L(α): Calculate the gradient of the Lagrangian function L(α) with respect to and The gradient of , the Lagrangian function L(α) is defined as in, is the kernel function, which is used to calculate the similarity between two samples. and The gradient of , Here λ is the inverse of the regularization parameter C, that is, λ = 1 / C; Solve the quadratic programming problem: Simplify the problem into a quadratic programming problem to find the new and values, which will maximize L(α) under the constraints. The quadratic programming problem is expressed as , where the elements of the B matrix are It is related to the kernel function and the label. c is a constant. Solving this quadratic programming problem can get the new and The value of renew and : Updates with the new value found by the solver and , the updated value must satisfy the constraints of SVM, which are expressed as ; Calculate the new value of b: and After that, a support vector is selected (i.e. or samples greater than 0) to calculate the new value of b, , where w is the weight vector, expressed as , select a support vector , so that >0, then Substitute the expression for w into the above formula to get the new value of b; Check KKT conditions, check updated , Do a and b satisfy the KKT condition. If not, continue iterating; if they do, the algorithm converges and finds the best separating hyperplane; S423: Model evaluation: After training is completed, use the test dataset to evaluate the performance of the model; S424: Preliminary classification: Input the new data into the trained support vector machine model to obtain preliminary classification results.

[0024] In one embodiment, for the above S43, in S43, the specific steps of the random forest refinement classification are: S431: Feature selection: Select features useful for classification from the preliminary classification results and personalized evaluation indicators, and select the feature with the smallest Gini impurity as the feature useful for classification. The Gini impurity is expressed as ,in, is the number of categories, is the probability of the i-th category; S432: Construct decision tree: For each tree in the random forest, a training set with the same size as the original data set is extracted from the training data set using the Bootstrap method. At each split point, a part of the features is randomly selected for inspection, and the optimal feature is selected to split the node based on the Gini impurity. The above process is repeated until the stopping condition is met; S433: Ensemble decision: build multiple trees, each of which is trained independently. For new samples to be classified, use each decision tree in the random forest to classify them, and decide the final category of the new sample by majority voting; S434: Input preliminary classification results: The preliminary classification results are used as input data and input into the trained random forest model; S435: Obtain refined classification results: The random forest model outputs refined classification results based on the characteristics of the input data.

[0025] An artificial intelligence-based school curriculum quality assessment system used in an artificial intelligence-based school curriculum quality assessment method includes: Data collection module: responsible for collecting data related to school curriculum quality assessment; Data processing module: responsible for pre-processing the collected data and processing and analyzing the data related to the school curriculum quality assessment according to the evaluation dimensions; Model evaluation module: responsible for using support vector machines to perform preliminary classification of processed and analyzed data. Based on the preliminary classification of the support vector machine, the random forest is used in combination with personalized evaluation indicators to refine the classification results. Result interpretation and application module: responsible for interpreting and explaining the evaluation results, and making targeted course improvement suggestions based on the evaluation results.

[0026] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A school curriculum quality assessment method based on artificial intelligence, characterized in that: The following steps are involved: S1: Determine the evaluation dimensions and personalized evaluation indicators: The evaluation dimensions include the curriculum education orientation, curriculum content structure, academic quality standards, and curriculum implementation and effects. The personalized evaluation indicators include disciplines and grades. S2: Data collection: Collect data based on evaluation dimensions and personalized evaluation indicators; S3: Data processing: Process and analyze the collected data according to the evaluation dimensions; S4: Data evaluation: Use support vector machine to preliminarily classify the data processed and analyzed in S3. Based on the preliminary classification by support vector machine, use random forest and combine personalized evaluation indicators to refine the classification results and obtain the course quality evaluation report of the target school. S5: Interpretation of results and application suggestions: Explain and illustrate the evaluation results, and make targeted course improvement suggestions based on the evaluation results.

2. According to claim 1, a school curriculum quality assessment method based on artificial intelligence is characterized in that: The collected data are processed and analyzed according to the educational orientation of the course. Through content analysis, it is checked whether the core literacy training objectives are clearly put forward in the course, and the specific manifestation of the training objectives in the course is observed. Key information and indicators are extracted. Through student surveys or tests, students' feedback on the course content is collected, their understanding and acceptance are evaluated, and a student feedback report is formed.

3. The method for evaluating school curriculum quality based on artificial intelligence according to claim 1, characterized in that: The collected data are processed and analyzed according to the course content structure. By analyzing whether the course content strengthens the connection with the educational goals, in the arrangement of the course content, the internal connection between each knowledge point and the corresponding relationship between each knowledge point and the educational goals are analyzed.

4. The method for evaluating school curriculum quality based on artificial intelligence according to claim 1, characterized in that: The collected data is processed and analyzed according to the academic quality standards. By collecting and analyzing the statistical data of students' academic performance and core literacy, an evaluation report on the implementation effect of the academic quality standards is formed. By comparing the academic quality standards and course objectives, their consistency is evaluated. The consistency is expressed as ,in, Indicates consistency, Indicates academic performance, represents the course objective, and n represents the number of evaluation samples.

5. The method for evaluating school curriculum quality based on artificial intelligence according to claim 1, characterized in that: The collected data is processed and analyzed based on the course implementation and effects. Through student satisfaction surveys, students' feedback and evaluation on the course implementation effects are collected. By collecting and analyzing statistical data on students' academic performance and core literacy improvement, the role of course implementation in students' all-round development is evaluated.

6. The method for evaluating school curriculum quality based on artificial intelligence according to claim 1, characterized in that: In S4, the specific steps of the data evaluation are: S41: Data preparation and processing: Receive the data processed and analyzed in S3, and extract key features and labels from the processed and analyzed data; S42: Support vector machine preliminary classification: The extracted key features and labels are input into the trained support vector machine model for preliminary classification; S43: Random forest refined classification: Use the preliminary classification results and personalized evaluation indicators as training data to train the random forest model, and input the preliminary classification results into the trained random forest model for refined classification; S44: Derive the evaluation report: Integrate the results of the support vector machine preliminary classification and the random forest refined classification to form a complete course quality evaluation report.

7. The method for evaluating school curriculum quality based on artificial intelligence according to claim 6, characterized in that: In S42, the specific steps of the support vector machine preliminary classification are: S421: Input data: input the extracted key features and labels into the support vector machine model; S422: Learning algorithm: The support vector machine model uses SMO to find the best separation hyperplane, the separation hyperplane is in the form of ,in, is the weight vector, is the input feature vector, It is a bias term that makes samples of different categories as correctly classified as possible; S423: Model evaluation: After training is completed, use the test dataset to evaluate the performance of the model; S424: Preliminary classification: Input the new data into the trained support vector machine model to obtain preliminary classification results.

8. The method for evaluating school curriculum quality based on artificial intelligence according to claim 6, characterized in that: In S43, the specific steps of the random forest refinement classification are: S431: Feature selection: Select features useful for classification from the preliminary classification results and personalized evaluation indicators, and select the feature with the smallest Gini impurity as the feature useful for classification. The Gini impurity is expressed as ,in, is the number of categories, is the probability of the i-th category. The value of Gini impurity is between 0 and 1. The smaller the value, the purer the data set, that is, the more samples belong to the same category. The larger the value, the more impure the data set, that is, the more uniform the distribution of sample categories. S432: Construct decision tree: For each tree in the random forest, a training set with the same size as the original data set is extracted from the training data set using the Bootstrap method. At each split point, a part of the features is randomly selected for inspection, and the optimal feature is selected to split the node based on the Gini impurity. The above process is repeated until the stopping condition is met; S433: Ensemble decision: build multiple trees, each of which is trained independently. For new samples to be classified, use each decision tree in the random forest to classify them, and decide the final category of the new sample by majority voting; S434: Input preliminary classification results: The preliminary classification results are used as input data and input into the trained random forest model; S435: Obtain refined classification results: The random forest model outputs refined classification results based on the characteristics of the input data.

9. The artificial intelligence-based school curriculum quality assessment system used in the artificial intelligence-based school curriculum quality assessment method according to claim 1, characterized in that: include: Data collection module: responsible for collecting data related to school curriculum quality assessment; Data processing module: responsible for pre-processing the collected data and processing and analyzing the data related to the school curriculum quality assessment according to the evaluation dimensions; Model evaluation module: responsible for using support vector machines to perform preliminary classification of processed and analyzed data. Based on the preliminary classification of the support vector machine, the random forest is used in combination with personalized evaluation indicators to refine the classification results. Result interpretation and application module: responsible for interpreting and explaining the evaluation results, and making targeted course improvement suggestions based on the evaluation results.

Images