Data acquisition and analysis method and system based on artificial intelligence

Abstract

The invention discloses a data collection and analysis method and system based on artificial intelligence, and belongs to the technical field of education systems.The data collection and analysis method based on artificial intelligence comprises the following steps that S1, data collection is conducted, specifically, fixed factor data and random factor data of students and data indicating whether the corresponding students pass an examination or not are collected, the fixed factor data comprises learning behavior data, historical score data and learning participation degree data, and the random factor data comprises teacher teaching methods and class average scores; s2, data processing: carrying out data processing on the fixed factor data and the random factor data, wherein the data processing comprises data cleaning, normalization and feature selection; and S3, model training: taking the processed fixed factor data and random factor data as input, taking data indicating whether the students pass the examination as output, and constructing a mixed effect model. The method has the advantage that the data accuracy is improved by integrating the fixed factors and the random factors.

Description

Technical Field

[0001] The present invention belongs to the technical field of educational systems, and in particular, relates to a data collection and analysis method and system based on artificial intelligence. Background Art

[0002] In teaching practice, using artificial intelligence to predict whether a student can pass an exam can help teachers and students better understand their learning progress and mastery. Artificial intelligence systems can analyze a large amount of data such as students' learning behavior, historical grades, participation, and test results to identify learning patterns and potential areas of difficulty. Such analysis can provide students with personalized feedback and suggestions to help them improve their learning methods in a targeted manner and increase the possibility of passing the exam. At the same time, teachers can also use this information to adjust teaching strategies to ensure that all students receive the necessary support.

[0003] In the prior art, when collecting and analyzing student data, the collected student data is relatively idealized, and usually only fixed factor data of students are considered, such as learning behavior, historical grades, and learning participation. These are data that students can change through hard work, but the influence of random factors is not considered, such as teacher teaching methods, class average scores, etc. This leads to inaccurate information collection, analysis and prediction results. Summary of the invention

[0004] In response to the problems existing in the prior art, the present invention provides a data collection and analysis method and system based on artificial intelligence, which has the advantage of improving data accuracy by integrating fixed factors and random factors, and solves the problems of the prior art.

[0005] The present invention is implemented in this way: the data collection and analysis method based on artificial intelligence comprises the following steps: Step S1, collecting data: collecting students' fixed factor data, random factor data and data on whether the corresponding students have passed the exam. The fixed factor data includes learning behavior data, historical performance data, and learning participation data. The random factor data includes teacher teaching methods and class average scores. Step S2, data processing: performing data processing on the fixed factor data and the random factor data, including data cleaning, normalization, and feature selection; Step S3, model training: taking the processed fixed factor data and random factor data as input and taking the data of whether the students passed the exam as output, a mixed effect model is constructed, wherein the mixed effect model is: Yi = Xiβ + Ziui + εi; in: Yi is the observation result of the ith student; (e.g., whether he passed the exam) Xi is the fixed factor design matrix, β is the fixed factor parameter vector; Zi is the random factor design matrix, ui is the random factor parameter vector; εi is the random error term; Step S4, data prediction: After the model training is completed, new student data is predicted.

[0006] As a preferred embodiment of the present invention, in step S1, the step of collecting the fixed factor data, random factor data and data of whether the student has passed the examination includes: Step S11, determining the type of collected data; Step S12, determining the time range and quantity of collected data; Step S13, collecting data in the database through a data collector.

[0007] As preferred embodiments of the present invention, the types of collected data are: The learning behavior data includes: class study time, after-school self-study time, and cram school time; The historical performance data include: average data of in-class tests; The learning engagement data includes: the number of interactions and the number of completed assignments; The teacher's teaching method is the teacher's teaching score; The class average score includes the average score data of the corresponding course class test.

[0008] As a preferred embodiment of the present invention, collecting the data of whether the corresponding student has passed the examination includes: Collect students’ test scores; Get the passing score data of the corresponding exam; If the test score is less than the passing score, it is judged that the test has not been passed; otherwise, it is judged that the test has been passed.

[0009] As a preferred embodiment of the present invention, the data processing includes: Data cleaning: remove incomplete and erroneous data and add new data to the database to keep the number of data groups unchanged; Data normalization: Scaling data to a uniform range or distribution.

[0010] As a preferred embodiment of the present invention, in step S3, the fixed factor design matrix Xi is obtained by converting the fixed factor data into numerical form and organizing it into a matrix form, and the specific steps are as follows: Step S311. Convert each observation value of each fixed factor into numerical data, wherein each observation value of the fixed factor is class study time, after-school self-study time, cram school time, historical performance data, number of interactions, and number of completed homework; Step S312: Encode or convert the numerical data according to the requirements of the prediction model, and convert the categorical data into virtual variables.

[0011] Step S313. Organize the converted data into a matrix by columns, where each column represents a fixed factor and each row represents an observed object.

[0012] Step S314: Estimate the fixed factor parameter vector β by maximum likelihood estimation or restricted maximum likelihood estimation.

[0013] As preferred embodiment of the present invention, the steps of constructing the random factor design matrix Zi are as follows: Step S321. Determine the source of random factors as the teacher’s teaching method and the average score of class history grades.

[0014] Step S322. Assign an identifier to each random factor: assign corresponding codes to different teachers' teaching methods, and assign codes to different class history grade averages; Step S323. Construct a matrix: Construct a random factor design matrix Zi based on these identifiers, where each column corresponds to a random factor and each row corresponds to an observation value. If a random factor is applicable to a student, fill in 1 in the corresponding position of the matrix, otherwise fill in 0.

[0015] As a preferred embodiment of the present invention, the determination of the random factor parameter vector ui includes: Step S324. Estimating random factor parameters: In the model training phase, the value of the random factor parameter vector ui is estimated by statistical methods; Step S325. Apply the model: Once the model training is completed, for new student data, the known random factor design matrix Zi and the estimated random factor parameter vector ui are combined with the fixed factor design matrix Xi and the fixed factor parameter vector β, and the mixed effect model formula Yi=Xiβ+Ziui+εi is used to predict whether the student has passed the exam.

[0016] The data acquisition and analysis system based on artificial intelligence includes a central control module, and the central control module is signal-connected to a data acquisition module, a data processing module, a prediction module and an application module; The data collection module is used to collect students' fixed factor data, random factor data and data on whether the corresponding students have passed the exam. The fixed factor data includes learning behavior data, historical performance data, and learning participation data. The random factor data includes teacher teaching methods and class average scores. The data processing module is used to process the fixed factor data and the random factor data, including data cleaning, normalization, and feature selection; The prediction module is used to take the processed fixed factor data and random factor data as input, and take the data of whether the student passed the exam as output to construct a mixed effect model, wherein the mixed effect model is: Yi = Xiβ + Ziui + εi; in: Yi is the observation result of the ith student; (e.g., whether he passed the exam) Xi is the fixed factor design matrix, β is the fixed factor parameter vector; Zi is the random factor design matrix, ui is the random factor parameter vector; εi is the random error term; The application module is used to predict new student data after the model training is completed.

[0017] As a preferred embodiment of the present invention, a data type selection subunit is used to determine the type of collected data; The data range selection subunit is used to determine the time range and quantity of the collected data; The collection subunit is used to collect data in the database through a data collector.

[0018] The types of collected data are: The learning behavior data includes: class study time, after-school self-study time, and cram school time; The historical performance data include: average data of in-class tests; The learning engagement data includes: the number of interactions and the number of completed assignments; The teacher's teaching method is the teacher's teaching score; The class average score includes the average score data of the corresponding course class test.

[0019] Compared with the prior art, the present invention has the following beneficial effects: The present invention constructs a mixed effect model by comprehensively considering fixed factors and random factors, which can more accurately analyze and predict students' learning effects, provide more valuable information for students and teachers, and thus improve the quality of education and teaching effects. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 It is a flowchart of a data collection and analysis method based on artificial intelligence provided by an embodiment of the present invention; Figure 2 is a flowchart of data collection provided by an embodiment of the present invention; Figure 3 It is a flowchart of constructing a fixed factor design matrix Xi and calculating a fixed factor parameter vector β provided by an embodiment of the present invention.

[0021] Figure 4 It is a flowchart of constructing a random factor design matrix Zi provided by an embodiment of the present invention; Figure 5 It is a flowchart of determining the random factor parameter vector ui provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0022] In order to further understand the content, features and effects of the present invention, the following embodiments are given as examples and described in detail with reference to the accompanying drawings.

[0023] The structure of the present invention is described in detail below in conjunction with the accompanying drawings.

[0024] See also Figure 1-Figure 5 The data collection and analysis method based on artificial intelligence provided by the embodiment of the present invention comprises the following steps: Step S1, data collection: collect students' fixed factor data, random factor data and corresponding data on whether students have passed the exam. The fixed factor data include learning behavior data, historical performance data, and learning participation data, which reflect the learning performance that students can change through hard work. The random factor data include teacher teaching methods and class average scores, which may affect students' learning effects but are not completely controlled by students.

[0025] Step S2, data processing: performing data processing on the fixed factor data and the random factor data, including data cleaning, normalization, and feature selection; Step S3, model training: taking the processed fixed factor data and random factor data as input and taking the data of whether the students passed the exam as output, a mixed effect model is constructed, wherein the mixed effect model is: Yi = Xiβ + Ziui + εi; in: Yi is the observation result of the ith student; (e.g., whether he passed the exam) Xi is the fixed factor design matrix, β is the fixed factor parameter vector; Zi is the random factor design matrix, ui is the random factor parameter vector; εi is the random error term; Step S4, data prediction: After the model training is completed, new student data is predicted to provide students with personalized feedback and suggestions.

[0026] By introducing artificial intelligence data collection and analysis methods, this solution can more comprehensively consider the factors that affect students' learning outcomes, thereby improving the accuracy of information collection, analysis and prediction. Specifically, this solution solves the problems existing in the existing technology through the following steps: 1. Data collection: Not only fixed factor data of students, such as learning behavior, historical grades and learning participation, but also random factor data, such as teacher teaching methods and class average scores, are collected. Although these random factors are not completely controlled by students, they have a significant impact on learning outcomes.

[0027] 2. Data processing: Perform data cleaning, normalization and feature selection on the collected fixed factor data and random factor data to ensure data quality and provide an accurate basis for subsequent analysis.

[0028] 3. Model construction: Construct a mixed effect model, which not only contains fixed factor parameters, but also introduces random factor parameters, which can simultaneously consider the impact of fixed factors and random factors on students' learning effects. The model formula is Yi=Xiβ+Ziui+εi, where Yi represents the observation results of the i-th student, Xi is the fixed factor design matrix, β is the fixed factor parameter vector, Zi is the random factor design matrix, ui is the random factor parameter vector, and εi is the random error term.

[0029] 4. Model training and prediction: By training the mixed effect model, students’ test results can be predicted. After the model training is completed, the model can be used to predict new student data, thereby providing personalized feedback and suggestions to students.

[0030] 5. Adjustment of teaching strategies: By analyzing the prediction results of the model, teachers can understand which random factors have a greater impact on students’ learning outcomes, and adjust teaching strategies accordingly to ensure that all students receive the necessary support.

[0031] Exemplarily, in step S1, the step of collecting the fixed factor data, random factor data and data on whether the student has passed the examination includes: Step S11, determining the type of collected data; Step S12, determining the time range and quantity of collected data; Step S13, collecting data in the database through a data collector.

[0032] Exemplarily, the types of collected data are: The learning behavior data includes: class study time, after-school self-study time, and cram school time; The historical performance data include: average data of in-class tests; The learning engagement data includes: the number of interactions and the number of completed assignments; The teacher's teaching method is the teacher's teaching score; The class average score includes the average score data of the corresponding course class test.

[0033] Exemplarily, collecting data on whether the corresponding student has passed the exam includes: Collect students’ test scores; Get the passing score data of the corresponding exam; If the test score is less than the passing score, it is judged that the test has not been passed; otherwise, it is judged that the test has been passed.

[0034] Exemplarily, the data processing includes: Data cleaning: remove incomplete and erroneous data and add new data to the database to keep the number of data groups unchanged; Data normalization: Scaling data to a uniform range or distribution.

[0035] Exemplarily, in step S3, the fixed factor design matrix Xi is obtained by converting the fixed factor data into numerical form and organizing it into a matrix form, and the specific steps are as follows: Step S311. Convert each observation value of each fixed factor into numerical data, wherein each observation value of the fixed factor is class study time, after-school self-study time, cram school time, historical performance data, number of interactions, and number of completed homework; Step S312: Encode or convert the numerical data according to the requirements of the prediction model, and convert the categorical data into virtual variables (dummy variables).

[0036] Step S313. Organize the converted data into a matrix by columns, where each column represents a fixed factor and each row represents an observed object (eg, a student).

[0037] Step S314: Estimate the fixed factor parameter vector β by maximum likelihood estimation (MLE) or restricted maximum likelihood estimation (REML).

[0038] For example, first, data is collected. For example, for class study time, there are the following observations: 3 hours, 2.5 hours, 3.5 hours, etc. For historical performance data, there are scores: 85 points, 90 points, 88 points, etc. For categorical data, such as class type, it may be converted into a dummy variable. For example, if the class type is divided into three categories: A, B, and C, three dummy variables are created, each corresponding to a class type. If the student belongs to this type, the variable value is 1, otherwise it is 0.

[0039] Next, organize these numerical data into the form of a matrix Xi. For example, if there is a student whose data is 3 hours of class study, 1 hour of after-school self-study, 2 hours of cram school, 85 points in history, 10 times of interaction, and 5 times of homework completion, then in the matrix, the row of this student is as follows: [3,1,2,85,10,5] Each column represents a fixed factor and each row represents a student.

[0040] The fixed factor parameter vector β is estimated from the data using statistical methods. For example, if the maximum likelihood estimation (MLE) is used, the statistical software will find the best β value based on the data and model settings through an optimization algorithm to maximize the probability of the observed data.

[0041] Assuming that statistical software is used for the analysis, the software automatically performs these steps and outputs the final fixed factor parameter vector β. For example, the software outputs: β=[0.2,0.15,0.1,0.05,0.02,0.03] This represents the average effect of each fixed factor on academic performance. For example, the first element of β, 0.2, may mean that for every additional hour of class time, the expected academic performance will increase by an average of 0.2 points. Through such a model, we can better understand the impact of different factors on students' academic performance.

[0042] Exemplarily, the steps of constructing the random factor design matrix Zi are as follows: Step S321. Determine the source of random factors as the teacher’s teaching method and the average score of class history grades.

[0043] Step S322. Assign an identifier to each random factor: assign corresponding codes to different teachers' teaching methods, and assign codes to different class history grade averages; Step S323. Construct a matrix: Construct a random factor design matrix Zi based on these identifiers, where each column corresponds to a random factor and each row corresponds to an observation (i.e., a student). If a random factor (such as a specific teacher's teaching method) is applicable to a student, fill in 1 in the corresponding position of the matrix, otherwise fill in 0.

[0044] Exemplarily, the determination of the random factor parameter vector ui includes: Step S324. Estimate random factor parameters: During the model training phase, the values ​​of the random factor parameter vector ui are estimated by statistical methods (such as maximum likelihood estimation or Bayesian estimation). These parameters represent the contribution of the random factor to the model output.

[0045] Step S325. Apply the model: Once the model training is completed, for new student data, the known random factor design matrix Zi and the estimated random factor parameter vector ui are combined with the fixed factor design matrix Xi and the fixed factor parameter vector β, and the mixed effect model formula Yi=Xiβ+Ziui+εi is used to predict whether the student has passed the exam.

[0046] In this way, teachers can adjust their teaching strategies based on the personalized feedback and suggestions provided by the model to ensure that all students receive the necessary support.

[0047] Exemplarily, the following example is used to illustrate how to construct the random factor design matrix Zi and the random factor parameter vector ui.

[0048] Suppose there is a mixed effects model with two random factors: teacher teaching method (Method) and class history average score (ClassAvg). There are three teachers (A, B, C) and two classes (1, 2). Each class has 5 students.

[0049] First, construct the random factor design matrix Zi: Determine the sources of random factors: teacher teaching method (Method) and class history average score (ClassAvg).

[0050] Assign an identifier to each random factor: teachers A, B, and C are represented by 1, 2, and 3 respectively; classes 1 and 2 are represented by 1 and 2 respectively.

[0051] Construct the matrix: For each student, create a matrix with two columns corresponding to the two random factors.

[0052] For example, for students in class 1, the random factor design matrix Zi is as follows: Student Method ClassAvg 1 1 1 2 0 1 3 1 1 4 1 1 5 0 1 For students in class 2, the matrix is ​​as follows:

[0053] Next, determine the random factor parameter vector ui: Estimation of random factor parameters: Assume that the following random factor parameter vector ui is estimated by statistical methods: Method ClassAvg 0.3 0.5 -0.2 -0.3 0.1 0.1 Teacher A's teaching method has a positive effect on student performance (0.3) and the average history score of Class 1 (0.5), while Teacher B's teaching method has a slight negative effect (-0.2) and a slight negative effect (-0.3) on the average history score of Class 2.

[0054] It should be noted that the random error term εi is calculated by the difference between the observed data and the model predicted value in the mixed effect model. Specifically, the random error term represents the part of the variation that the model cannot explain, that is, the difference between the observed value Yi and the model predicted value Xiβ+Ziui. In statistics, random error terms are usually assumed to be independent and identically distributed random variables with zero mean and a certain variance. During the model training process, the model parameters β and ui are estimated by minimizing some measure of the error term (such as the sum of squares in the least squares method). In practical applications, the calculation of random error terms is usually done automatically by statistical software or machine learning libraries.

[0055] The data acquisition and analysis system based on artificial intelligence includes a central control module, and the central control module is signal-connected to a data acquisition module, a data processing module, a prediction module and an application module; The data collection module is used to collect students' fixed factor data, random factor data, and data on whether the corresponding students have passed the exam. The fixed factor data includes learning behavior data, historical performance data, and learning participation data, which reflect the learning performance that students can change through hard work. The random factor data includes the teacher's teaching method and the class average score. These factors may affect the students' learning effect, but are not completely controlled by the students.

[0056] The data processing module is used to process the fixed factor data and the random factor data, including data cleaning, normalization, and feature selection; The prediction module is used to take the processed fixed factor data and random factor data as input, and take the data of whether the student passed the exam as output to construct a mixed effect model, wherein the mixed effect model is: Yi = Xiβ + Ziui + εi; in: Yi is the observation result of the ith student; (e.g., whether he passed the exam) Xi is the fixed factor design matrix, β is the fixed factor parameter vector; Zi is the random factor design matrix, ui is the random factor parameter vector; εi is the random error term; The application module is used to predict new student data after the model training is completed, so as to provide personalized feedback and suggestions to students.

[0057] Wherein, the data acquisition module includes: A data type selection subunit is used to determine the type of collected data; The data range selection subunit is used to determine the time range and quantity of the collected data; The collection subunit is used to collect data in the database through a data collector.

[0058] The types of collected data are: The learning behavior data includes: class study time, after-school self-study time, and cram school time; The historical performance data include: average data of in-class tests; The learning engagement data includes: the number of interactions and the number of completed assignments; The teacher's teaching method is the teacher's teaching score; The class average score includes the average score data of the corresponding course class test.

[0059] Working principle of the present invention: When used, fixed factor data refers to data that is the same for all observations, while random factor data refers to data that varies between different observations. Students' grades may be affected by the average grades of their classes, and the average grades of each class may be affected by the teaching methods of the class teacher. In this case, the students' learning data can be used as fixed effects, and the class and teacher are random effects. Combining the two types of data can improve the accuracy of data analysis.

[0060] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0061] 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 data collection and analysis method based on artificial intelligence, characterized in that: The following steps are involved: Step S1, collecting data: collecting students' fixed factor data, random factor data and data on whether the corresponding students have passed the exam. The fixed factor data includes learning behavior data, historical performance data, and learning participation data. The random factor data includes teacher teaching methods and class average scores. Step S2, data processing: performing data processing on the fixed factor data and the random factor data, including data cleaning, normalization, and feature selection; Step S3, model training: taking the processed fixed factor data and random factor data as input and taking the data of whether the students passed the exam as output, a mixed effect model is constructed, wherein the mixed effect model is: Yi = Xiβ + Ziui + εi; in: Yi is the observation result of the i-th student; Xi is the fixed factor design matrix, β is the fixed factor parameter vector; Zi is the random factor design matrix, ui is the random factor parameter vector; εi is the random error term; Step S4, data prediction: After the model training is completed, new student data is predicted.

2. The data collection and analysis method based on artificial intelligence according to claim 1, characterized in that: In step S1, the steps of collecting the fixed factor data, random factor data and data on whether the student has passed the examination include: Step S11, determining the type of collected data; Step S12, determining the time range and quantity of collected data; Step S13, collecting data in the database through a data collector.

3. The data collection and analysis method based on artificial intelligence as claimed in claim 2, characterized in that: The types of collected data are: The learning behavior data includes: class study time, after-school self-study time, and cram school time; The historical performance data include: average data of in-class tests; The learning participation data includes: the number of interactions and the number of completed assignments; The teacher's teaching method is the teacher's teaching score; The class average score includes the average score data of the corresponding course class test.

4. The data collection and analysis method based on artificial intelligence according to claim 2, characterized in that: The data of whether the corresponding student has passed the examination includes: Collect students’ test scores; Get the passing score data of the corresponding exam; If the test score is less than the passing score, it is judged that the test has not been passed; otherwise, it is judged that the test has been passed.

5. The data collection and analysis method based on artificial intelligence according to claim 1, characterized in that: The data processing includes: Data cleaning: remove incomplete and erroneous data and add new data to the database to keep the number of data groups unchanged; Data normalization: Scaling data to a uniform range or distribution.

6. The artificial intelligence-based data collection and analysis method according to claim 4, characterized in that: In step S3, the fixed factor design matrix Xi is obtained by converting the fixed factor data into numerical form and organizing it into a matrix form. The specific steps are as follows: Step S311. Convert each observation value of each fixed factor into numerical data, wherein each observation value of the fixed factor is class study time, after-school self-study time, cram school time, historical performance data, number of interactions, and number of completed homework; Step S312. Encode or convert the numerical data according to the requirements of the prediction model, and convert the categorical data into dummy variables; Step S313. Organize the converted data into a matrix by columns, where each column represents a fixed factor and each row represents an observed object; Step S314: Estimate the fixed factor parameter vector β by maximum likelihood estimation or restricted maximum likelihood estimation.

7. The data collection and analysis method based on artificial intelligence according to claim 6, characterized in that: The steps to construct the random factor design matrix Zi are as follows: Step S321. Determine the source of random factors as the teacher's teaching method and the average score of class history grades; Step S322. Assign an identifier to each random factor: assign corresponding codes to different teachers' teaching methods, and assign codes to different class history grade averages; Step S323. Construct a matrix: Construct a random factor design matrix Zi based on these identifiers, where each column corresponds to a random factor and each row corresponds to an observation value. If a random factor is applicable to a student, fill in 1 in the corresponding position of the matrix, otherwise fill in 0.

8. The data collection and analysis method based on artificial intelligence according to claim 7, characterized in that: The determination of the random factor parameter vector ui includes: Step S324. Estimating random factor parameters: In the model training phase, the value of the random factor parameter vector ui is estimated by statistical methods; Step S325. Apply the model: Once the model training is completed, for new student data, the known random factor design matrix Zi and the estimated random factor parameter vector ui are combined with the fixed factor design matrix Xi and the fixed factor parameter vector β, and the mixed effect model formula Yi=Xiβ+Ziui+εi is used to predict whether the student has passed the exam.

9. The data acquisition and analysis system based on artificial intelligence includes a central control module, characterized in that: The central control module is signal-connected to a data acquisition module, a data processing module, a prediction module and an application module; The data collection module is used to collect students' fixed factor data, random factor data and data on whether the corresponding students have passed the exam. The fixed factor data includes learning behavior data, historical performance data, and learning participation data. The random factor data includes teacher teaching methods and class average scores. The data processing module is used to process the fixed factor data and the random factor data, including data cleaning, normalization, and feature selection; The prediction module is used to take the processed fixed factor data and random factor data as input, and take the data of whether the student passed the exam as output to construct a mixed effect model, wherein: The mixed effects model is: Yi = Xiβ + Ziui + εi; in: Yi is the observation result of the i-th student; Xi is the fixed factor design matrix, β is the fixed factor parameter vector; Zi is the random factor design matrix, ui is the random factor parameter vector; εi is the random error term; The application module is used to predict new student data after the model training is completed.

10. The artificial intelligence-based data collection and analysis system according to claim 9, characterized in that: The data acquisition module comprises: A data type selection subunit is used to determine the type of collected data; The data range selection subunit is used to determine the time range and quantity of the collected data; A collection subunit, used for collecting data in a database through a data collector; The types of collected data are: The learning behavior data includes: class study time, after-school self-study time, and cram school time; The historical performance data include: average data of in-class tests; The learning participation data includes: the number of interactions and the number of completed assignments; The teacher's teaching method is the teacher's teaching score; The class average score includes the average score data of the corresponding course class test.

Images