AI-driven comprehensive evaluation system for learning ability of primary and secondary school students

Abstract

The invention discloses a primary and secondary school student learning ability comprehensive evaluation system based on AI driving, and the system comprises an information input module which is used for recording the basic information of students; the material management module interacts with the information input module and is used for centrally storing digital teaching resources and basic information of students; the evaluation module interacts with the information input module and the material management module and is used for learning progress and testing according to the digital teaching resources; the learning analysis module interacts with the evaluation module and provides error analysis and learning suggestions according to the student learning result evaluation result; and the man-machine interaction module is used for establishing interaction between a user and the comprehensive evaluation system. According to the method, the accuracy and individuation of evaluation are improved, learning weaknesses of students can be more accurately identified through automatic scoring and error analysis realized by the AI technology, and customized training questions are provided according to the weaknesses, so that the learning efficiency is improved.

Description

Technical Field

[0001] The present invention belongs to the field of assessment, and more specifically, it relates to an AI-driven comprehensive assessment system for the learning ability of primary and secondary school students. Background Art

[0002] In the modern education system, the assessment of students' learning ability has become an important means to improve the quality of education. With the advancement of science and technology, especially the development of artificial intelligence (AI) technology, an AI-driven comprehensive assessment system for the learning ability of primary and secondary school students has emerged, aiming to provide students with personalized and accurate learning assessment and guidance through high-tech means. This system uses AI technology to conduct a comprehensive analysis of students' learning behavior, grades and cognitive abilities, so as to tailor learning plans and improvement strategies for each student.

[0003] Traditional learning ability assessment methods usually rely on paper tests and teachers' subjective judgment, which is not only time-consuming and labor-intensive, but also difficult to achieve personalization and dynamic tracking. With the development of information technology, especially the application of AI technology, AI-driven comprehensive learning ability assessment systems have become possible. These systems can automatically collect students' learning data, analyze students' learning status through algorithm models, and provide real-time feedback on learning suggestions, greatly improving the efficiency and accuracy of assessments.

[0004] However, the existing technology still has shortcomings:

[0005] Most existing AI comprehensive evaluation systems use fixed weighting coefficients in weighted calculations. This method is feasible in most cases, but the output results are not accurate in special cases. For example, a student has raised the accuracy of basic questions to more than 90% by doing a large number of questions, while the accuracy of comprehensive questions and extension questions is only 20%. If the weighted calculation is performed by the AI ​​comprehensive evaluation system in the existing technology, the final result of the output will evaluate the student as an average or excellent student. In fact, the student only memorized the solution to similar questions by doing a large number of questions, and did not understand the knowledge points, which will lead to misjudgment of the AI ​​comprehensive evaluation system. Therefore, we propose an AI-driven comprehensive evaluation system for the learning ability of primary and secondary school students. Summary of the invention

[0006] The purpose of the present invention is to solve the shortcomings existing in the prior art and propose an AI-driven comprehensive evaluation system for the learning ability of primary and secondary school students. The system adopts a primary evaluation and a secondary evaluation method, and flexibly assigns weights according to the actual correct answer rate, so that the results obtained are more accurate and effectively solve the problems raised in the background technology.

[0007] To achieve the above object, the present invention provides the following technical solutions:

[0008] The AI-driven comprehensive assessment system for the learning ability of primary and secondary school students includes:

[0009] An information entry module is used to record basic information of students. The information entry module has a built-in permission allocation unit, which is used to provide different access rights and function interfaces according to user roles;

[0010] A material management module interacts with the information entry module and is used to centrally store digital teaching resources and basic information of students. The material management module has a built-in retrieval unit, which is used to classify teaching resources;

[0011] An evaluation module interacts with the information entry module and the material management module, and is used to evaluate the student learning results according to the learning progress and test of the digital teaching resources, and match the evaluation results with the basic information of the students in the information entry module. It has a built-in AI unit, and is used to automatically generate personalized test questions according to the evaluation results of the student learning results, which are used for targeted training of students. The personalized test questions are similar questions generated according to the wrong questions in the evaluation results;

[0012] A learning analysis module interacts with the evaluation module, provides error analysis and learning suggestions based on the evaluation results of students’ learning outcomes, and uses machine learning algorithms to evaluate students’ learning abilities and mastery. The learning analysis module has a built-in tracking unit for tracking and recording students’ learning behaviors, monitoring students’ learning progress, and identifying learning problems in a timely manner.

[0013] The human-computer interaction module is used to establish interaction between the user and the comprehensive evaluation system.

[0014] Preferably, the human-computer interaction module includes:

[0015] A PC host, wherein the information entry module, the material management module, the evaluation module and the learning analysis module are all integrated inside the PC host, and the display end of the PC host is used to display the learning analysis module and the output results of the learning analysis module;

[0016] It also includes a detection module worn on the student's body, which is wirelessly connected to the PC host for real-time monitoring of the student's physical condition when the student is studying or evaluating, judging the student's physical condition through the student's heart rate status, judging whether the student needs rest based on the student's physical condition, and adjusting the teaching and exercise plan based on the heart rate status.

[0017] Preferably, the information entry module is used to provide a user interface, and collects basic information of the user manually or by data import through the human-computer interaction module, including name, age, gender, school, and grade. The information entry module includes:

[0018] The permission allocation unit is used to dynamically allocate system access rights and function interfaces according to user roles;

[0019] The data verification subunit is used to ensure that the entered data is accurate and conforms to the data standards and formats set by the system;

[0020] The security protection subunit uses modern encryption algorithms and security protocols to protect stored data from unauthorized access;

[0021] The data synchronization subunit is used to ensure data consistency and synchronization between the information entry module and the material management module, the assessment module, and the learning analysis module;

[0022] The user information management subunit enables students, teachers and parents to access and manage student information, but different data and operation options are displayed according to permissions;

[0023] The log subunit records all user operations for system troubleshooting.

[0024] Preferably, the material management module includes:

[0025] The data storage layer is used to store all digital teaching resources, including textbooks, exercises, video tutorials, interactive exercises, and basic information about students;

[0026] The data processing layer is used to manage and organize the above data, including indexing, retrieving, updating and protecting the data. The built-in retrieval unit of the data processing layer can efficiently classify the teaching resources by subject and difficulty level, so as to facilitate quick access;

[0027] The interface layer provides interactive interfaces with the information entry module and the evaluation module to ensure that data can be smoothly transmitted between different modules;

[0028] The material management module also allows authorized users to upload new teaching resources or update existing resources. The uploaded resources need to go through a review process to ensure the quality and applicability of the content, and the built-in retrieval unit can automatically or manually add keyword tags, subject classifications and difficulty levels to each teaching resource to improve the searchability of the resources.

[0029] Preferably, the specific workflow of the evaluation module is:

[0030] Information collection: collecting students’ basic information and learning progress data from the information entry module and material management module;

[0031] One-time assessment: Based on the students' learning progress, online tests are conducted regularly. The tests include multiple-choice questions, fill-in-the-blank questions, and application questions. After the test is completed, the system automatically scores and analyzes the types and frequencies of students' errors.

[0032] Information matching: matching the assessment results with the students’ basic information and recording them in the students’ historical learning files;

[0033] Targeted exercises: The built-in AI unit uses machine learning algorithms to automatically generate targeted exercises based on students’ wrong questions and ability levels.

[0034] Secondary assessment: students use practice questions for targeted training and the system scores them twice;

[0035] Output the results, perform weighted calculation on the first and second evaluations, and output the final results.

[0036] Preferably, in the targeted exercises, a progressive question-asking method is adopted, and the expression is:

[0037] L X =(J c ×a)+(Z h ×b)+(T z ×c);

[0038] Where, L X All exercises in Targeted Practice, J c Basic questions are based on the knowledge content of this chapter learned in the book; Z h Comprehensive questions are based on the knowledge learned in this chapter and the knowledge of other chapters. Comprehensive questions also include questions of the same type as wrong questions in an assessment. z This is an extension question. It is based on the easy-to-make mistakes in the knowledge of this chapter learned in the book and the wrong knowledge points in the previous assessment. a, b, c are the proportion coefficients of various types of questions, and a+b+c=100%;

[0039] When outputting the results, the first and second evaluation results are weighted, and then the average is taken to obtain the result. The output expression of the result is:

[0040] S final = ;

[0041] In the formula, S1 is the score of the first evaluation, S2 is the score of the second evaluation, and S final Weighted average score, i.e. the result output, w1 and w2 are weighted values ​​respectively, and w1+w2=1;

[0042] In the above formula, the values ​​of w1 and w2 are not fixed. After obtaining w1 through the accuracy of S1, the value of w2 is calculated through w1. The calculation formula for the value of w1 is:

[0043] w1=100%-S1 accuracy;

[0044] The value of W2 is calculated as follows:

[0045] w2=1-w1;

[0046] The above calculation method avoids the situation where the output result value is high due to the accuracy of an assessment being too high, thereby preventing inaccurate judgment of students' learning results.

[0047] Preferably, the specific process of using the machine learning algorithm in the learning analysis module to evaluate the student's learning ability and mastery level is as follows:

[0048] Collect students' test scores, homework scores, and classroom interaction records on different knowledge points, clean the data to handle missing values ​​and outliers, and convert non-numeric data into a format suitable for model processing;

[0049] Extract meaningful features from the raw data, including the average test score of each knowledge point and the type of incorrect questions. According to business needs, classify students' learning status into different categories, including not mastered, basically mastered, and fully mastered.

[0050] Use machine learning algorithms to build an initial model and optimize model performance by adjusting parameters;

[0051] Use the trained model to make predictions on the students’ new learning data and evaluate the accuracy of the model.

[0052] Preferably, the initial model expression constructed using a machine learning algorithm is:

[0053] ;

[0054] In the formula, p k is the proportion of samples in the kth category in the data set, and the data set is the evaluation results of students’ learning outcomes; |y| is the total number of categories;

[0055] According to the above formula, we can get:

[0056] ;

[0057] Where a represents a discrete attribute, V is the number of values ​​of attribute a, and D v is a subset divided according to the vth value of attribute a, |D| is the size of data set D, that is, the total number of samples, |D v ∣ is a subset D v Size, G ini (D) is the Gini index of the data set D, which measures the impurity of the data. ini_index (D,a) is the weighted Gini index of the data set after partitioning according to attribute a.

[0058] Preferably, the detection module comprises:

[0059] Sensors are used to collect students’ acceleration, heart rate, and body temperature parameters during exercise;

[0060] The processor is used to perform preliminary analysis and processing on the collected data and issue a warning in time if any abnormality is found;

[0061] Wireless communication module, used to send data to the host PC;

[0062] The display unit is used for teachers and parents to obtain students' physical fitness monitoring data, view various assessment reports, understand students' physical conditions, and adjust teaching and exercise plans based on feedback.

[0063] Technical effects and advantages of the present invention: Compared with traditional technologies, the AI-driven comprehensive evaluation system for learning ability of primary and secondary school students provided by the present invention has the following advantages:

[0064] Improve the accuracy and personalization of assessments. Automatic scoring and error analysis achieved through AI technology can more accurately identify students' learning weaknesses and provide customized training questions based on these weaknesses, thereby improving learning efficiency.

[0065] Dynamically adjust the learning plan. The system can dynamically adjust the teaching content and difficulty according to the students' real-time learning progress and results to ensure that the learning plan always meets the students' actual needs;

[0066] Promote students' active learning. Through instant feedback and personalized suggestions, students can more clearly understand their learning status, stimulate their interest in learning and motivation for self-improvement;

[0067] Strengthen the connection between home and school. Parents can intuitively understand their children's learning progress and existing problems through the system, which helps parents better support their children's learning and communicate with teachers in a timely manner;

[0068] Improve the monitoring and analysis capabilities of learning outcomes. The system's built-in learning analysis module can track students' learning behaviors and outcomes over a long period of time, providing educational researchers with a large amount of real data, which helps optimize teaching methods and strategies.

[0069] Management of adaptive teaching resources. The material management module allows for rapid retrieval and updating of educational resources, ensuring the timeliness and relevance of teaching content;

[0070] Enhance physical health monitoring. Through the detection module, students' physical conditions, such as heart rate, can be monitored, and students can be reminded to rest when necessary to prevent excessive fatigue and protect their health. BRIEF DESCRIPTION OF THE DRAWINGS

[0071] Figure 1This is the architecture diagram of the AI-driven comprehensive evaluation system for learning ability of primary and secondary school students of the present invention;

[0072] Figure 2 4 is a flowchart of the evaluation module in an embodiment of the present invention. DETAILED DESCRIPTION

[0073] In order to make the purpose, technical scheme and advantages of the present invention clearer, the present invention is further described in detail below in conjunction with specific embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0074] The present invention provides an AI-driven comprehensive assessment system for the learning ability of primary and secondary school students, which aims to solve the various deficiencies in the existing education assessment system. The system adopts a refined method of primary and secondary assessment, combined with dynamic weight adjustment based on the correctness of students' answers, in order to obtain more accurate and personalized assessment results. This method effectively overcomes the problems of over-reliance on subjective judgment, low efficiency and lack of personalized teaching in traditional assessment.

[0075] On the technical level, the AI-driven comprehensive assessment system for primary and secondary school students' learning ability uses artificial intelligence technology, including machine learning and deep learning, to instantly score and assess students' responses. In this way, the system is able to identify the specific needs of each student and provide them with tailored learning materials and exercises, thereby improving learning efficiency and promoting in-depth understanding of knowledge.

[0076] In addition, the AI-driven comprehensive assessment system for primary and secondary school students’ learning ability also includes a multifunctional material management module, which can efficiently store and classify digital teaching resources to ensure that these resources can be easily accessed and used by teachers and students. This not only improves the utilization rate of educational resources, but also ensures the timeliness and relevance of teaching content.

[0077] The AI-driven comprehensive assessment system for primary and secondary school students' learning ability focuses on data security and user privacy protection, and uses modern encryption algorithms and strict data protection measures to prevent unauthorized data access and leakage. At the same time, the design of the human-computer interaction module focuses on user experience, allowing teachers, students and parents to easily interact with the system, obtain the required information, and provide feedback.

[0078] During the teaching process, the system of the present invention can also provide real-time feedback and suggestions based on the students' learning status, helping teachers adjust their teaching strategies, while also allowing parents to better understand their children's learning progress. The comprehensive performance and highly automated functions of the system greatly reduce the workload of teachers, allowing them to devote more energy to improving the quality of teaching.

[0079] like Figure 1 , an AI-driven comprehensive assessment system for the learning ability of primary and secondary school students, including:

[0080] The information entry module is used to record the basic information of students. The information entry module has a built-in permission allocation unit, which is used to provide different access rights and function interfaces according to user roles. The information entry module is used to provide a user interface. The human-computer interaction module collects basic user information manually or through data import, including name, age, gender, school, and grade. The information entry module includes:

[0081] The permission allocation unit is used to dynamically allocate system access rights and function interfaces according to user roles;

[0082] The data verification subunit is used to ensure that the entered data is accurate and conforms to the data standards and formats set by the system;

[0083] The security protection subunit uses modern encryption algorithms and security protocols to protect stored data from unauthorized access;

[0084] The data synchronization subunit is used to ensure data consistency and synchronization between the information entry module and the material management module, the assessment module, and the learning analysis module;

[0085] The user information management subunit enables students, teachers and parents to access and manage student information, but different data and operation options are displayed according to permissions;

[0086] The log subunit records all user operations for system troubleshooting.

[0087] The material management module interacts with the information entry module and is used to centrally store digital teaching resources and basic information of students. The material management module has a built-in retrieval unit, which is used to classify teaching resources. The material management module includes:

[0088] The data storage layer is used to store all digital teaching resources, including textbooks, exercises, video tutorials, interactive exercises, and basic information about students;

[0089] The data processing layer is used to manage and organize the above data, including data indexing, retrieval, updating and protection. The built-in retrieval unit of the data processing layer can efficiently classify teaching resources by subject and difficulty level for quick access;

[0090] The interface layer provides interactive interfaces with the information entry module and the evaluation module to ensure that data can be smoothly transmitted between different modules;

[0091] The material management module also allows authorized users to upload new teaching resources or update existing resources. The uploaded resources need to go through a review process to ensure the quality and applicability of the content, and the built-in retrieval unit can automatically or manually add keyword tags, subject classifications and difficulty levels to each teaching resource to improve the searchability of the resources.

[0092] The evaluation module interacts with the information entry module and the material management module, and is used to evaluate the students' learning results according to the learning progress and test of digital teaching resources, and match the evaluation results with the basic information of students in the information entry module. It has a built-in AI unit, which is used to automatically generate personalized test questions based on the evaluation results of students' learning results, and is used for targeted training of students. The personalized test questions are similar questions generated based on the wrong questions in the evaluation results;

[0093] like Figure 2 ,The specific workflow of the evaluation module is as follows:

[0094] Information collection: collecting students’ basic information and learning progress data from the information entry module and material management module;

[0095] One-time assessment: Based on the students' learning progress, online tests are conducted regularly. The tests include multiple-choice questions, fill-in-the-blank questions, and application questions. After the test is completed, the system automatically scores and analyzes the types and frequencies of students' errors.

[0096] Information matching: matching the assessment results with the students’ basic information and recording them in the students’ historical learning files;

[0097] Targeted exercises: The built-in AI unit uses machine learning algorithms to automatically generate targeted exercises based on students’ wrong questions and ability levels. In targeted exercises, a progressive question-setting method is used, and the expression is:

[0098] L X =(J c ×a)+(Z h ×b)+(T z ×c);

[0099] Where, L X All exercises in Targeted Practice, J cBasic questions are based on the knowledge content of this chapter learned in the book; Z h Comprehensive questions are based on the knowledge learned in this chapter and the knowledge of other chapters. Comprehensive questions also include questions of the same type as wrong questions in an assessment. z This is an extension question. It is based on the easy-to-make mistakes in the knowledge of this chapter learned in the book and the wrong knowledge points in the previous assessment. a, b, c are the proportion coefficients of various types of questions, and a+b+c=100%;

[0100] Secondary assessment: students use practice questions for targeted training and the system scores them twice;

[0101] Output the result, perform weighted calculation on the first and second evaluations, and output the final result; when outputting the result, weight the first and second evaluation results, and then take the average value to get the result. The output expression of the result is:

[0102] S final = ;

[0103] In the formula, S1 is the score of the first evaluation, S2 is the score of the second evaluation, and S final Weighted average score, i.e. the result output, w1 and w2 are weighted values ​​respectively, and w1+w2=1;

[0104] In the above formula, the values ​​of w1 and w2 are not fixed. After obtaining w1 through the accuracy of S1, the value of w2 is calculated through w1. The calculation formula for the value of w1 is:

[0105] w1=100%-S1 accuracy;

[0106] The value of W2 is calculated as follows:

[0107] w2=1-w1;

[0108] The above calculation method avoids the situation where the output result value is high due to the accuracy of an assessment being too high, thereby preventing inaccurate judgment of students' learning results.

[0109] For example, the above method is used to conduct comprehensive ability assessment on students A, B and C, and the specific scores are as follows:

[0110] One-time test score Secondary test score Student A 65 65 Student B 60 75 Student C 95 60

[0111] The following results are obtained from the above table:

[0112] Student A, the information collection and the first assessment score is 65 points, the targeted practice and the second assessment score is still 65 points, according to the score of the first assessment, calculate the value of w1: w1 = 100% - 65% = 35%, calculate the value of w2: w2 = 1 - w1 = 1 - 35% = 65%;

[0113] The final score S is calculated using the above weights final :

[0114] ;

[0115] Student B, the information collection and the first assessment score is 60 points, the targeted practice and the second assessment score is 75 points, according to the score of the first assessment, calculate the value of w1: w1 = 100% - 60% = 40%, calculate the value of w2: w2 = 1 - w1 = 1 - 40% = 60%;

[0116] The final score S is calculated using the above weights final :

[0117] ;

[0118] Student C, the information collection and the first assessment score is 95 points, the targeted practice and the second assessment score is 65 points, according to the score of the first assessment, calculate the value of w1: w1 = 100% - 60% = 40%, calculate the value of w2: w2 = 1 - w1 = 1 - 40% = 60%;

[0119] The final score S is calculated using the above weights final :

[0120] ;

[0121] The output results of the three students are as follows:

[0122] Primary evaluation score and weight assignment Secondary evaluation scores and weight assignment Output results of comprehensive assessment of learning ability Student A 65×35% 65×65% 65 Student B 60×40% 75×60% 69 Student C 95×5% 60×95% 61.75

[0123] Through the above examples, we can see how the AI-driven comprehensive assessment system for primary and secondary school students' learning ability can be actually applied to the learning situations of different students, and how to dynamically adjust the scoring weights according to the results of each assessment to obtain a more accurate assessment of learning ability. The core advantage of the system lies in its flexibility and adaptability, which can adjust the assessment criteria according to the actual performance of students, thereby providing more fair and personalized assessment results;

[0124] By analyzing the results of the first and second assessments of different students, the system can identify each student's learning status and areas that need improvement, and provide personalized learning suggestions accordingly. For example, in the case of student C, although the initial score was high, the second assessment score was low, indicating that there may be difficulties in understanding certain key concepts. The system can then provide more exercises in a targeted manner;

[0125] As shown by the improvement in student B’s performance, the system not only detects learning progress, but also adjusts the weights to ensure that the evaluation results truly reflect the student’s actual performance. This dynamic adjustment helps teachers and parents grasp students’ learning progress more accurately;

[0126] For students like Student A who have stable performance, the consistent scores provided by the system can serve as encouragement, allowing them to see the results of their continuous efforts, thereby increasing their motivation to learn;

[0127] The example of student C shows that high initial scores may be followed by learning fatigue or comprehension difficulties, and systematic assessment can detect such problems in time and solve them through subsequent practice and coaching;

[0128] By accurately assessing students' needs, the system helps educators allocate teaching resources and time more effectively, provide teaching tailored to students' specific needs, and improve teaching efficiency;

[0129] This AI-driven comprehensive learning ability assessment system not only improves the accuracy of assessment and the quality of education, but also promotes the all-round development of students and the effective use of educational resources through personalized learning suggestions and dynamic assessment mechanisms.

[0130] The learning analysis module interacts with the assessment module, provides error analysis and learning suggestions based on the assessment results of students' learning outcomes, and uses machine learning algorithms to evaluate students' learning ability and mastery. The learning analysis module has a built-in tracking unit to track and record students' learning behaviors, monitor students' learning progress, and identify learning problems in a timely manner. The specific process of using machine learning algorithms to evaluate students' learning ability and mastery in the learning analysis module is as follows:

[0131] Collect students' test scores, homework scores, and classroom interaction records on different knowledge points, clean the data to handle missing values ​​and outliers, and convert non-numeric data into a format suitable for model processing;

[0132] Extract meaningful features from the raw data, including the average test score of each knowledge point and the type of incorrect questions. According to business needs, classify students' learning status into different categories, including not mastered, basically mastered, and fully mastered.

[0133] Use machine learning algorithms to build an initial model, and optimize model performance by adjusting parameters; the expression for building an initial model using a machine learning algorithm is:

[0134] ;

[0135] In the formula, p k is the proportion of samples in the kth category in the data set, and the data set is the evaluation results of students’ learning outcomes; |y| is the total number of categories;

[0136] According to the above formula, we can get:

[0137] ;

[0138] Where a represents a discrete attribute, V is the number of values ​​of attribute a, and D v is a subset divided according to the vth value of attribute a, |D| is the size of data set D, that is, the total number of samples, |D v ∣ is a subset D v Size, G ini (D) is the Gini index of the data set D, which measures the impurity of the data. ini_index (D,a) is the weighted Gini index of the data set after partitioning according to attribute a;

[0139] Use the trained model to make predictions on the students’ new learning data and evaluate the accuracy of the model.

[0140] Furthermore, the above learning analysis module has the following effects:

[0141] Error analysis: It can identify common types of errors made by students during the learning process and provide targeted correction suggestions; Learning suggestions: Based on students' learning ability and mastery level, the system generates personalized learning plans and resource recommendations; Behavior tracking: The built-in tracking unit records students' learning behavior, monitors learning progress, and promptly identifies learning problems; Model optimization: By continuously collecting new learning data, the system can dynamically adjust model parameters to improve the accuracy of the evaluation.

[0142] Exemplarily, the operation of the built-in tracking unit of the learning analysis module is as follows:

[0143] There is a student D who is using an online learning platform integrated with the learning analysis module of the present invention. Student D is currently studying mathematics, especially algebra and geometry;

[0144] Tracking unit function

[0145] Behavior records: The tracking unit records the time when Student D logs into the platform each time, the chapters studied, the number of exercises completed, and the test scores.

[0146] Record the time Student D spends on each knowledge point, as well as the questions he raises and discussions he participates in in the discussion area.

[0147] Progress monitoring: The system automatically generates learning reports regularly to show student D's mastery of various knowledge points and his progress. The cycle for generating learning reports is one week.

[0148] By comparing the scores on each test, the system can identify areas in which Student D has made progress and areas that need improvement.

[0149] Problem discovery: If student D's accuracy rate for a certain knowledge point in several consecutive tests is always below a certain threshold, and the threshold is set as the passing line, that is, 60% accuracy rate, the system will mark this knowledge point as requiring attention.

[0150] The system also analyzes the type of error made by Student D, such as calculation errors, confusion of concepts, or incorrect problem-solving steps.

[0151] Feedback provision: Based on student D’s learning behavior and progress, the system provides personalized learning suggestions, such as recommending relevant video explanations, exercises or reference materials.

[0152] If it detects a decline in learning efficiency or a lack of motivation, the system will remind students to take a break or try different learning methods.

[0153] Based on the above, if student D often makes mistakes when solving algebraic equations, the tracking unit records this situation and finds through analysis that he mainly makes mistakes when moving terms and combining like terms. The system then pushes several targeted tutorial videos and adds exercises on related question types.

[0154] When learning geometry, if Student D's interaction on the platform decreases and his test scores do not improve, the tracking unit detects this change and prompts the teacher and parents that Student D may have difficulty understanding geometry concepts. The teacher then arranges one-on-one tutoring to help Student D overcome the difficulties.

[0155] The human-computer interaction module is used to establish the interaction between the user and the comprehensive evaluation system. The human-computer interaction module includes:

[0156] The PC host, the information input module, the material management module, the evaluation module and the learning analysis module are all integrated inside the PC host, and the display end of the PC host is used to display the learning analysis module and the output results of the learning analysis module;

[0157] It also includes a detection module worn on the student's body, which is wirelessly connected to the PC host and is used to monitor the student's physical state in real time when the student is studying or evaluating, judge the student's physical state through the student's heart rate state, judge whether the student needs to rest according to the student's physical state, and adjust the teaching and exercise plan according to the heart rate state;

[0158] The detection module includes:

[0159] Sensors are used to collect students’ acceleration, heart rate, and body temperature parameters during exercise;

[0160] The processor is used to perform preliminary analysis and processing on the collected data and issue a warning in time if any abnormality is found;

[0161] Wireless communication module, used to send data to the PC host;

[0162] The display unit is used for teachers and parents to obtain students' physical fitness monitoring data, view various assessment reports, understand students' physical conditions, and adjust teaching and exercise plans based on feedback.

[0163] It can be concluded from the above that the present invention has the following advantages compared with the conventional technology:

[0164] Improve the accuracy and personalization of assessments. Automatic scoring and error analysis achieved through AI technology can more accurately identify students' learning weaknesses and provide customized training questions based on these weaknesses, thereby improving learning efficiency.

[0165] Dynamically adjust the learning plan. The system can dynamically adjust the teaching content and difficulty according to the students' real-time learning progress and results to ensure that the learning plan always meets the students' actual needs;

[0166] Promote students' active learning. Through instant feedback and personalized suggestions, students can more clearly understand their learning status, stimulate their interest in learning and motivation for self-improvement;

[0167] Strengthen the connection between home and school. Parents can intuitively understand their children's learning progress and existing problems through the system, which helps parents better support their children's learning and communicate with teachers in a timely manner;

[0168] Improve the monitoring and analysis capabilities of learning outcomes. The system's built-in learning analysis module can track students' learning behaviors and outcomes over a long period of time, providing educational researchers with a large amount of real data, which helps optimize teaching methods and strategies.

[0169] Management of adaptive teaching resources. The material management module allows for rapid retrieval and updating of educational resources, ensuring the timeliness and relevance of teaching content;

[0170] Enhance physical health monitoring. Through the detection module, students' physical conditions, such as heart rate, can be monitored, and students can be reminded to rest when necessary to prevent excessive fatigue and protect their health.

[0171] Finally, it should be noted that the above is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, it is still possible for those skilled in the art to modify the technical solutions described in the aforementioned embodiments, or to make equivalent substitutions for some of the technical features therein. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. An AI-driven comprehensive assessment system for the learning ability of primary and secondary school students, characterized by: include: An information entry module is used to record basic information of students. The information entry module has a built-in permission allocation unit, which is used to provide different access rights and function interfaces according to user roles; A material management module interacts with the information entry module and is used to centrally store digital teaching resources and basic information of students. The material management module has a built-in retrieval unit, which is used to classify teaching resources; An evaluation module interacts with the information entry module and the material management module, and is used to evaluate the student learning results according to the learning progress and test of the digital teaching resources, and match the evaluation results with the basic information of the students in the information entry module. It has a built-in AI unit, and is used to automatically generate personalized test questions according to the evaluation results of the student learning results, which are used for targeted training of students. The personalized test questions are similar questions generated according to the wrong questions in the evaluation results; A learning analysis module interacts with the evaluation module, provides error analysis and learning suggestions based on the evaluation results of students’ learning outcomes, and uses machine learning algorithms to evaluate students’ learning abilities and mastery. The learning analysis module has a built-in tracking unit for tracking and recording students’ learning behaviors, monitoring students’ learning progress, and identifying learning problems in a timely manner. The human-computer interaction module is used to establish interaction between the user and the comprehensive evaluation system.

2. According to claim 1, the AI-driven comprehensive evaluation system for learning ability of primary and secondary school students is characterized in that: The human-computer interaction module comprises: A PC host, wherein the information entry module, the material management module, the evaluation module and the learning analysis module are all integrated inside the PC host, and the display end of the PC host is used to display the learning analysis module and the output results of the learning analysis module; It also includes a detection module worn on the student's body, which is wirelessly connected to the PC host for real-time monitoring of the student's physical condition when the student is studying or evaluating, judging the student's physical condition through the student's heart rate status, judging whether the student needs rest based on the student's physical condition, and adjusting the teaching and exercise plan based on the heart rate status.

3. The AI-driven comprehensive evaluation system for learning ability of primary and secondary school students according to claim 1 is characterized in that: The information entry module is used to provide a user interface, and collects basic user information manually or through data import through the human-computer interaction module, including name, age, gender, school, and grade. The information entry module includes: The permission allocation unit is used to dynamically allocate system access rights and function interfaces according to user roles; The data verification subunit is used to ensure that the entered data is accurate and conforms to the data standards and formats set by the system; The security protection subunit uses modern encryption algorithms and security protocols to protect stored data from unauthorized access; The data synchronization subunit is used to ensure data consistency and synchronization between the information entry module and the material management module, the assessment module, and the learning analysis module; The user information management subunit enables students, teachers and parents to access and manage student information, but different data and operation options are displayed according to permissions; The log subunit records all user operations for system troubleshooting.

4. The AI-driven comprehensive evaluation system for learning ability of primary and secondary school students as described in claim 1 is characterized in that: The material management module includes: The data storage layer is used to store all digital teaching resources, including textbooks, exercises, video tutorials, interactive exercises, and basic information about students; The data processing layer is used to manage and organize the above data, including indexing, retrieving, updating and protecting the data. The built-in retrieval unit of the data processing layer can efficiently classify the teaching resources by subject and difficulty level, so as to facilitate quick access; The interface layer provides interactive interfaces with the information entry module and the evaluation module to ensure that data can be smoothly transmitted between different modules; The material management module also allows authorized users to upload new teaching resources or update existing resources. The uploaded resources need to go through a review process to ensure the quality and applicability of the content, and the built-in retrieval unit can automatically or manually add keyword tags, subject classifications and difficulty levels to each teaching resource to improve the searchability of the resources.

5. The AI-driven comprehensive evaluation system for learning ability of primary and secondary school students according to claim 1 is characterized in that: The specific workflow of the evaluation module is as follows: Information collection: collecting students’ basic information and learning progress data from the information entry module and material management module; One-time assessment: Based on the students' learning progress, online tests are conducted regularly. The tests include multiple-choice questions, fill-in-the-blank questions, and application questions. After the test is completed, the system automatically scores and analyzes the types and frequencies of students' errors. Information matching: matching the assessment results with the students’ basic information and recording them in the students’ historical learning files; Targeted exercises: The built-in AI unit uses machine learning algorithms to automatically generate targeted exercises based on students’ wrong questions and ability levels. Secondary assessment: students use practice questions for targeted training and the system scores them twice; Output the results, perform weighted calculation on the first and second evaluations, and output the final results.

6. The AI-driven comprehensive evaluation system for learning ability of primary and secondary school students according to claim 5 is characterized in that: In the targeted exercises, a progressive question-asking method is adopted, and the expression is: L X =(J c ×a)+(Z h ×b)+(T z ×c); Where, L X All exercises in Targeted Practice, J c Basic questions are based on the knowledge content of this chapter learned in the book; Z h Comprehensive questions are based on the knowledge learned in this chapter and the knowledge of other chapters. Comprehensive questions also include questions of the same type as wrong questions in an assessment. z This is an extension question. It is based on the easy-to-make mistakes in the knowledge of this chapter learned in the book and the wrong knowledge points in the previous assessment. a, b, c are the proportion coefficients of various types of questions, and a+b+c=100%; When outputting the results, the first and second evaluation results are weighted, and then the average is taken to obtain the result. The output expression of the result is: S final = ； In the formula, S1 is the score of the first evaluation, S2 is the score of the second evaluation, and S final Weighted average score, i.e. the result output, w1 and w2 are weighted values ​​respectively, and w1+w2=1; In the above formula, the values ​​of w1 and w2 are not fixed. After obtaining w1 through the accuracy of S1, the value of w2 is calculated through w1. The calculation formula for the value of w1 is: w1=100%-S1 accuracy; The value of W2 is calculated as follows: w2=1-w1; The above calculation method avoids the situation where the output result value is high due to the accuracy of an assessment being too high, thereby preventing inaccurate judgment of students' learning results.

7. The AI-driven comprehensive evaluation system for learning ability of primary and secondary school students according to claim 1 is characterized in that: The specific process of using machine learning algorithms to evaluate students' learning ability and mastery level in the learning analysis module is as follows: Collect students' test scores, homework scores, and classroom interaction records on different knowledge points, clean the data to handle missing values ​​and outliers, and convert non-numeric data into a format suitable for model processing; Extract meaningful features from the raw data, including the average test score of each knowledge point and the type of incorrect questions. According to business needs, classify students' learning status into different categories, including not mastered, basically mastered, and fully mastered. Use machine learning algorithms to build an initial model and optimize model performance by adjusting parameters; Use the trained model to make predictions on the students’ new learning data and evaluate the accuracy of the model.

8. The AI-driven comprehensive evaluation system for learning ability of primary and secondary school students according to claim 7 is characterized in that: The initial model expression constructed using the machine learning algorithm is: ； In the formula, p k is the proportion of samples in the kth category in the data set, and the data set is the evaluation results of students’ learning outcomes; |y| is the total number of categories; According to the above formula, we can get: ； Where a represents a discrete attribute, V is the number of values ​​of attribute a, and D v is a subset divided according to the vth value of attribute a, |D| is the size of data set D, that is, the total number of samples, |D v ∣ is a subset D v Size, G ini (D) is the Gini index of the data set D, which measures the impurity of the data. ini_index (D,a) is the weighted Gini index of the data set after partitioning according to attribute a.

9. The AI-driven comprehensive evaluation system for learning ability of primary and secondary school students according to claim 2 is characterized in that: The detection module comprises: Sensors are used to collect students’ acceleration, heart rate, and body temperature parameters during exercise; The processor is used to perform preliminary analysis and processing on the collected data and issue a warning in time if any abnormality is found; Wireless communication module, used to send data to the PC host; The display unit is used for teachers and parents to obtain students' physical fitness monitoring data, view various assessment reports, understand students' physical conditions, and adjust teaching and exercise plans based on feedback.

Images