Classroom evaluation auxiliary method and system, medium and terminal equipment

Abstract

The invention discloses a classroom evaluation auxiliary method and system, a medium and terminal equipment, and the method comprises the steps: collecting initial score parameters of students, carrying out the data cleaning of the score parameters, inputting the score parameters according to a preset format, adding timestamps, and uploading the score parameters to a cloud database; carrying out segmentation by adopting a recursive character segmentation device to obtain student document blocks; obtaining a student historical score data set, and constructing a data pair input preset comprehensive evaluation large model to obtain a personal score summary; collecting classroom performance images of students, converting the classroom performance images into corresponding behavior tags, and uploading the behavior tags to the comprehensive evaluation large model to obtain classroom seriousness coefficient scores; performing association analysis through a preset association analysis algorithm in combination with the personal score summary and the classroom performance data table to obtain comprehensive quality analysis; and obtaining corresponding comprehensive quality analysis according to the input instruction. By integrating data and analysis results of multiple dimensions, learning and growth conditions of students can be reflected more comprehensively.

Description

Technical Field

[0001] The present application relates to the field of educational effectiveness, and specifically to a classroom evaluation auxiliary method, system, medium and terminal device. Background Art

[0002] As a crucial component of campus education, student personal quality evaluation plays a crucial role in comprehensively and objectively reflecting students' learning and growth. This evaluation often encompasses multiple aspects, including academic performance and on-campus performance. However, existing technologies rely primarily on manual evaluation by educators, a time-consuming and labor-intensive process that is susceptible to subjective influences. The workload can even lead to students simply conducting self-evaluations and using these results as a comprehensive evaluation for the semester or year. This lacks a deep understanding of student personal quality evaluations. Summary of the Invention

[0003] This application proposes a classroom evaluation auxiliary method, system, medium and terminal device, which comprehensively reflects students' learning and growth by integrating data and analysis results from multiple dimensions.

[0004] A first aspect of the present application provides a classroom evaluation assistance method, the method comprising:

[0005] Collect students' initial score parameters, clean the score parameters and enter them into a preset format to obtain the first score file, add a timestamp to the first score file and upload it to the cloud database;

[0006] Using a recursive character splitter to split the first score file into student document blocks, wherein each student document block includes all relevant subject scores and wrong question types in the student's exam;

[0007] Obtain a dataset of students' historical performance and build a comprehensive evaluation model for data pair training to obtain individual performance summaries;

[0008] Collect students' classroom performance images in real time, convert them into corresponding behavior labels, and upload them to the comprehensive evaluation model to obtain classroom seriousness coefficient scores;

[0009] Combined with the individual performance summary and classroom performance data table, a correlation analysis is performed using a preset correlation analysis algorithm to obtain a comprehensive quality analysis;

[0010] According to the input instructions, you can obtain the corresponding comprehensive quality analysis.

[0011] The above scheme uses a comprehensive evaluation model to conduct a comprehensive analysis of students' academic performance and generate detailed performance reports. At the same time, it combines classroom performance data with deep correlation analysis to reveal the potential connection between academic performance and classroom behavior. Through comprehensive analytical methods, it more accurately depicts students' learning status and provides strong support for educational decision-making.

[0012] Furthermore, the step of obtaining the first score file is as follows:

[0013] The student's initial performance parameters include daily homework scores, test scores, and wrong question types;

[0014] Normalize the collected student initial score parameters and set relevant score thresholds to remove invalid data;

[0015] The first score file is set in a table format with student names as rows and subjects, wrong question types and scores as columns, and a timestamp is added to each row of data.

[0016] Furthermore, the data is constructed to input a preset comprehensive evaluation model to obtain a summary of individual performance, specifically:

[0017] The student historical performance data set includes historical student performance data and corresponding teacher comments;

[0018] Preprocess the historical performance dataset, using the format of subject: score: wrong question type - teacher's comments as input to train a comprehensive evaluation model.

[0019] Semantic analysis is used to perform part-of-speech tagging and extract keywords, and QLoRa is used to perform local fine-tuning on the comprehensive evaluation model;

[0020] Output personal performance summary through comprehensive evaluation model.

[0021] Furthermore, obtaining the classroom seriousness coefficient score includes:

[0022] Collect historical images related to the classroom and annotate relevant classroom behaviors;

[0023] The supervised training method enables comprehensive evaluation of the large model's ability to acquire action recognition;

[0024] The students' attitudes in class are obtained in real time during the period, and their class seriousness is quantified through the class seriousness coefficient.

[0025] Furthermore, the calculation formula for the classroom seriousness coefficient score is as follows:

[0026] S = w1×A+w2×P+w3×D;

[0027] Among them, A is the concentration, P is the classroom participation, D is the posture correction, and w1w2w3 are the corresponding weights respectively.

[0028] The above scheme comprehensively considers students' multi-dimensional performance in class by regularly calculating each student's classroom seriousness coefficient score, and ensures the fairness and accuracy of the evaluation results by adjusting the adaptive weights to meet the needs of different subjects and different teaching stages.

[0029] Furthermore, the association analysis is performed using a preset association analysis algorithm to obtain a comprehensive quality analysis, specifically:

[0030] Receive and merge the individual performance summary and class diligence coefficient scores output by the comprehensive evaluation model to obtain a comprehensive data set;

[0031] The preset correlation analysis algorithm is used to correlate and analyze the comprehensive data set, and the confidence level is calculated to obtain a comprehensive quality analysis and upload it to the cloud database.

[0032] The above scheme combines the summary of grades and the score of classroom attentiveness to conduct correlation analysis, evaluate the comprehensive quality of each student, and deeply explore the comprehensive quality of students; by exploring the correlation rules between grades and classroom performance, it provides data support for the adjustment of teaching strategies.

[0033] Furthermore, obtaining a corresponding comprehensive quality analysis according to the input instruction includes:

[0034] Receive specific input instructions and identify key information in the instructions through keyword matching;

[0035] According to the instruction analysis results, relevant comprehensive quality analysis is obtained from the cloud database.

[0036] This technical solution significantly reduces the workload of educators by combining a comprehensive evaluation model with multi-dimensional data analysis. At the same time, it relies on objective data to reduce interference from key factors, making evaluation results more realistic. By adjusting weights and algorithm parameters, it automatically adapts to the needs of different subjects and teaching stages. By integrating data and analysis results from multiple dimensions, it can more comprehensively reflect students' learning and growth, making evaluation results more accurate and reliable.

[0037] A second aspect of the present application provides a classroom evaluation assistance system, the system comprising: a score collection module, a score analysis module, a performance acquisition module, and a correlation analysis module;

[0038] The score collection module collects students' initial score parameters, cleans the score parameters and enters them in a preset format to obtain a first score file, adds a timestamp to the first score file and uploads it to the cloud database;

[0039] The performance analysis module uses a recursive character segmenter to segment the first performance file to obtain student document blocks, wherein each student document block includes scores of all relevant subjects and types of wrong questions in the student's examination;

[0040] Obtain a student's historical performance data set, build a data pair and input it into a preset comprehensive evaluation model to obtain a summary of individual performance;

[0041] The performance acquisition module collects students' classroom performance images in real time, converts them into corresponding behavior labels, and uploads them to the comprehensive evaluation model to obtain the classroom seriousness coefficient score;

[0042] The association analysis module combines the individual performance summary and the classroom performance data table to perform association analysis using a preset association analysis algorithm to obtain a comprehensive quality analysis;

[0043] According to the input instructions, you can obtain the corresponding comprehensive quality analysis.

[0044] A third aspect of the present application provides a storage medium storing computer-readable program code, which, when executed, implements the steps of a classroom evaluation assistance method described in any one of the embodiments of the present application.

[0045] The fourth aspect of the present application provides a terminal device, which includes: a terminal device including a processor and a memory, the memory storing a computer program, and the processor implementing the steps of a classroom evaluation assistance method described in any one of the embodiments of the present application when executing the computer program. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] In order to more clearly illustrate the technical solution of the present application, the following is a brief introduction to the drawings required for use in the implementation. Obviously, the drawings described below are only some implementation methods of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0047] Figure 1 This is a flowchart of a classroom evaluation assistance method provided by an embodiment of the present application;

[0048] Figure 2 This is a structural diagram of a classroom evaluation auxiliary system provided by an embodiment of the present application; DETAILED DESCRIPTION

[0049] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely for the purpose of explaining the present invention and are not intended to limit the present invention. In addition, the technical features involved in the various embodiments of the present invention described below may be combined with each other as long as they do not conflict with each other.

[0050] Example 1

[0051] See also Figure 1 As shown, a classroom evaluation auxiliary method provided by an embodiment of the present application includes steps S101-S104, which are specifically:

[0052] Step S101: Collecting students' initial score parameters, cleaning the score parameters and entering them into a preset format to obtain a first score file, adding a timestamp to the first score file and uploading it to a cloud database;

[0053] Furthermore, the step of obtaining the first score file is as follows:

[0054] The student's initial performance parameters include daily homework scores, test scores, and wrong question types;

[0055] Normalize the collected student initial score parameters and set relevant score thresholds to remove invalid data;

[0056] The first score file is set in a table format with student names as rows and subjects, wrong question types and scores as columns, and a timestamp is added to each row of data.

[0057] Specifically, students' regular homework scores, test scores, and types of wrong questions are collected. Each student is set as a separate row according to the student ID, and each subject and wrong question type is entered in a separate column format. The collected data is cleaned, score thresholds are set to remove invalid data, and the main types of wrong questions are classified; corresponding score thresholds are set according to actual teaching conditions, such as: in daily exams, the full score is set to 100 points. When data with a test score greater than 100 points or less than 0 points appears, it is determined to be invalid data and deleted; a timestamp is added to each row of data to mark the data collection time, so as to help educators observe the changing trends of students' grades over time, and compare the homework scores and test scores collected at different times for subsequent analysis.

[0058] Furthermore, a recursive character splitter is used to segment the first score file to obtain student document blocks, wherein each student document block includes scores of all relevant subjects and types of wrong questions in the student's examination;

[0059] It's important to note that the score data uploaded to the cloud database is segmented using a recursive character segmenter, with each line being segmented as a document chunk, ensuring that each student is represented as an independent document chunk. The recursive character segmenter accepts a character list as a parameter and segments the score data file into independent student document chunks. Each document chunk contains all relevant subject scores and error type data for a particular student's exam. The segmented student document chunks are then fed into the pre-set macro model.

[0060] Step S102: Obtain a student's historical performance data set, and construct a comprehensive evaluation model for data pair training to obtain a summary of individual performance;

[0061] Furthermore, the data is constructed to input a preset comprehensive evaluation model to obtain a summary of individual performance, specifically:

[0062] The student historical performance data set includes historical student performance data and corresponding teacher comments;

[0063] Preprocess the historical performance dataset, using the format of subject: score: wrong question type - teacher's comments as input to train a comprehensive evaluation model.

[0064] Semantic analysis is used to perform part-of-speech tagging and extract keywords, and QLoRa is used to perform local fine-tuning on the comprehensive evaluation model;

[0065] Get a summary of your personal performance through a comprehensive evaluation model.

[0066] In a specific embodiment, a historical performance data set and corresponding summary comments are collected to construct a historical data set. The collected data is preliminarily organized to create a file of each student's performance and comments. For example, "[Chinese: 89: Reading Comprehension, Mathematics: 92: Function, English: 71: Word Selection, Physics: 90: Mechanics Comprehensive Questions, Chemistry: 85: Chemical Calculation Questions, Biology: 75: Genetic Map Analysis Questions, History: 33: Material Analysis Questions, Geography: 21: Chart Questions, Politics: 10: Essay Questions]," and the summary comments are: "You performed very well in science, especially in mathematics and physics, with high scores of 92 and 90 respectively, showing that you are good in these two areas." You possess a solid foundation of knowledge and good logical thinking skills. Your scores in Chemistry and Biology were also quite good, at 85 and 75 respectively, indicating that you do a good job of understanding complex concepts and applying scientific principles. However, your performance in the humanities appears to require more attention and improvement. Your scores in History, Geography, and Politics were relatively low, at 33, 21, and 10, respectively. This may indicate that you have difficulty memorizing, understanding, and analyzing humanities knowledge, or perhaps you haven't found the right study methods. We recommend that you adjust your study strategies to strengthen your understanding and retention of humanities subjects while maintaining your strengths in science. You can improve your humanities performance through reading, discussion, and writing exercises.

[0067] Natural language processing tools were used to perform part-of-speech tagging and word segmentation on teacher comments. Based on the part-of-speech tagging results and named entity recognition, keywords reflecting learning performance, strengths, and weaknesses were extracted from the comments. Stop words were removed to further streamline the keyword list. Fine-tuning was performed using QLoRa. The preprocessed historical performance dataset was divided into training and validation sets to train the comprehensive evaluation model. By adding a low-rank matrix to the model parameters to adapt to the task, the number of parameter updates during training was reduced, thereby accelerating training and reducing the risk of overfitting. During training, the model's performance on the validation set was monitored, and parameters were adjusted accordingly until optimal performance was achieved. The new performance data, in the preprocessed format, was input into the trained comprehensive evaluation model to generate the corresponding individual performance summary.

[0068] Step S103: Collect students’ classroom performance images in real time, convert them into corresponding behavior labels, and upload them to the comprehensive evaluation model to obtain classroom seriousness coefficient scores;

[0069] Furthermore, obtaining the classroom seriousness coefficient score includes:

[0070] Collect historical images related to the classroom and annotate relevant classroom behaviors;

[0071] The supervised training method enables comprehensive evaluation of the large model's ability to acquire action recognition;

[0072] The students' attitudes in class are obtained in real time during the period, and their class seriousness is quantified through the class seriousness coefficient.

[0073] It should be noted that students' behaviors in class are collected in real time as their performance in class, and the behaviors are converted into corresponding behavioral labels through the big model and stored in the cloud database. By fine-tuning the big model, a large number of classroom-related historical images are collected, and various behaviors in the classroom are manually labeled, the big model is equipped with action recognition capabilities based on supervised training. In a specific embodiment, at the beginning of the class, the postures of students in the classroom within the cycle are obtained, and the students are associated with the corresponding classroom seriousness coefficient scores through face recognition. A classroom data table is created based on the name / ID of each student, the field name is set to each subject, and the column name is set to the date of each day. After obtaining the students' classroom performance in each subject on that day, the big model quantifies it into a classroom seriousness coefficient score.

[0074] Furthermore, the calculation formula for the classroom seriousness coefficient score is as follows:

[0075] S = w1×A+w2×P+w3×D;

[0076] Among them, A is the concentration, P is the classroom participation, D is the posture correction, and w1w2w3 are the corresponding weights respectively.

[0077] Specifically, the degree of concentration is obtained by (class duration - total duration of line of sight deviation) / class duration; participation is obtained by identifying students' specific actions, including raising hands, taking notes, answering questions, etc., and setting weights for each specific action, such as: the weight of raising hands to ask questions is 3, the weight of answering questions is 2, the weight of taking notes is 1, etc. The number of times students perform each action in class is counted and the total weight is calculated. The formula for obtaining participation is: total weight / (class duration × number of action types); posture correction is obtained by detecting students' sitting or standing posture through early annotation training of the model, and the calculation formula is: posture correction time / class duration; by quantifying the serious morning reading in class, the data is transmitted to the cloud data center and associated with the student's corresponding classroom data table.

[0078] Step S104: combining the individual performance summary and the classroom performance data table, performing a correlation analysis using a preset correlation analysis algorithm to obtain a comprehensive quality analysis;

[0079] According to the input instructions, you can obtain the corresponding comprehensive quality analysis.

[0080] Furthermore, the association analysis is performed using a preset association analysis algorithm to obtain a comprehensive quality analysis, specifically:

[0081] Receive and merge the individual performance summary and class diligence coefficient scores output by the comprehensive evaluation model to obtain a comprehensive data set;

[0082] The preset correlation analysis algorithm is used to correlate and analyze the comprehensive data set, and the confidence level is calculated to obtain a comprehensive quality analysis and upload it to the cloud database.

[0083] Specifically, the performance summary and classroom data table output by the large model are obtained from the cloud database, and a correlation analysis algorithm is used to perform correlation analysis on the student's performance and classroom attentiveness. In one specific embodiment, the FP-Growth correlation analysis algorithm is used to perform correlation analysis on the student's performance summary (including grades, error types, and performance summary) and the student's classroom data table (including subject, date, and classroom attentiveness quantitative score). The student's performance summary and classroom data table are merged according to the student's ID and subject to obtain a data set containing grades, error types, teacher comments and recommendations, and classroom attentiveness quantitative score. Each student's data is regarded as a transaction. For example, the mathematics transaction of student S001 can be {Math score: 94; Stage comments and recommendations: Good math logic, a little careless on simple questions; Main error type: Multiple choice questions; Classroom attentiveness quantitative score: 21}. The FP-Growth correlation analysis algorithm is combined with the set transaction item form, and a minimum support threshold is set. The data set is scanned to count the frequency of each item, and a frequent item set and FP tree are constructed. By recursively mining the FP tree, all frequent item sets are found, and then the association rules between different data items are discovered and the corresponding confidence levels are calculated to reveal the association between the performance summary and the quantitative value of classroom attentiveness.

[0084] Furthermore, obtaining a corresponding comprehensive quality analysis according to the input instruction includes:

[0085] Receive specific input instructions and identify key information in the instructions through keyword matching;

[0086] According to the instruction analysis results, relevant comprehensive quality analysis is obtained from the cloud database.

[0087] Specifically, teachers can input specific commands to generate corresponding output reports. For example, if you input the command "Evaluate each student's academic performance this semester," the output will include each student's test scores, main error types, and corresponding recommended comment keywords. If you input "Evaluate each student's classroom performance this month," the output will include a classroom data table for each student during that period. If you input "Output a comprehensive quality evaluation based on each student's academic performance and classroom performance this semester," the output will include a comprehensive quality analysis of each student based on the correlation analysis results, including grades, main error types, recommended comment keywords, classroom data table, and correlation analysis results. Teachers can obtain comprehensive quality evaluations of each student in batches through output, greatly improving their work efficiency and providing a more objective evaluation of each student's performance in school.

[0088] The beneficial effects of the present invention are: comprehensive analysis of students' academic performance is carried out through a large model, and in-depth correlation analysis is carried out using the large model output in combination with classroom performance data to reveal the potential connection between performance and classroom behavior. The comprehensive analysis method can more accurately depict students' learning status and provide strong support for educational decision-making; at the same time, by regularly calculating each student's classroom seriousness coefficient score, comprehensively considering the students' multi-dimensional performance in the classroom, and adaptively adjusting the weights through the large model to meet the needs of different subjects and different teaching stages; the student's stage-by-stage comprehensive quality report is adaptively output through the time segment of the teacher's input instruction, and the student's performance, main wrong question types, summary comment recommendations and classroom seriousness coefficient score in the corresponding time period are obtained through the setting of the timestamp, which facilitates the acquisition of students' stage-by-stage learning results and results analysis, and assists teachers in specifying stage-by-stage learning plans for students.

[0089] Example 2

[0090] Please refer to Figure 2 , a classroom evaluation auxiliary system provided by an embodiment of the present application, including: including: a score collection module, a score analysis module, a performance acquisition module and a correlation analysis module;

[0091] The score collection module collects students' initial score parameters, cleans the score parameters and enters them in a preset format to obtain a first score file, adds a timestamp to the first score file and uploads it to the cloud database;

[0092] The performance analysis module uses a recursive character segmenter to segment the first performance file to obtain student document blocks, wherein each student document block includes scores of all relevant subjects and types of wrong questions in the student's examination;

[0093] Obtain a dataset of students' historical performance and build a comprehensive evaluation model for data pair training to obtain individual performance summaries;

[0094] The performance acquisition module collects students' classroom performance images in real time, converts them into corresponding behavior labels, and uploads them to the comprehensive evaluation model to obtain the classroom seriousness coefficient score;

[0095] The association analysis module combines the individual performance summary and the classroom performance data table to perform association analysis using a preset association analysis algorithm to obtain a comprehensive quality analysis;

[0096] According to the input instructions, you can obtain the corresponding comprehensive quality analysis.

[0097] The above-mentioned classroom evaluation auxiliary system can implement the classroom evaluation auxiliary method of the above-mentioned method embodiment. The optional options in the above-mentioned method embodiment are also applicable to this embodiment and will not be described in detail here. The rest of the embodiments of this application can refer to the contents of the above-mentioned method embodiment. In certain preferred embodiments, no further description will be given.

[0098] Example 3

[0099] Accordingly, the present application also provides a computer-readable storage medium, which includes a stored computer program, wherein when the computer program is running, the device where the computer-readable storage medium is located is controlled to execute the classroom evaluation assistance method described in any one of the above embodiments.

[0100] Exemplarily, the computer program may be divided into one or more modules / units, which are stored in the memory and executed by the processor to complete the present application. The one or more modules / units may be a series of computer program instruction segments capable of performing specific functions, and the instruction segments are used to describe the execution process of the computer program in the terminal device.

[0101] The terminal device may be a computing device such as a desktop computer, a notebook computer, a PDA, a cloud server, etc. The terminal device may include, but is not limited to, a processor and a memory.

[0102] The processor may be a central processing unit (CPU), other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor, etc. The processor is the control center of the terminal device, connecting various parts of the entire terminal device using various interfaces and lines.

[0103] The memory can be used to store the computer programs and / or modules. The processor realizes various functions of the terminal device by running or executing the computer programs and / or modules stored in the memory and calling the data stored in the memory. The memory can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, at least one application required for a function, etc.; the data storage area can store data created according to the use of the mobile terminal, etc. In addition, the memory can include a high-speed random access memory and can also include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart memory card (SmartMedia Card, SMC), a secure digital (SecureDigital, SD) card, a flash card (Flash Card), at least one disk storage device, a flash memory device, or other volatile solid-state storage devices.

[0104] Wherein, if the module / unit integrated in the terminal device is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present application implements all or part of the process in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. Wherein, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form, etc. The computer-readable medium may include: any entity or device that can carry the computer program code, recording medium, U disk, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electrical carrier signal, telecommunication signal and software distribution medium, etc.

[0105] The above embodiments are merely descriptions of preferred embodiments of the present invention and are not intended to limit the scope of the present invention. Without departing from the design spirit of the present invention, various modifications and improvements made to the technical solutions of the present invention by ordinary engineering technicians in this field should fall within the scope of protection determined by the claims of the present invention.

Claims

1. A classroom evaluation auxiliary method, characterized in that: include: Collect students' initial score parameters, clean the score parameters and enter them into a preset format to obtain the first score file, add a timestamp to the first score file and upload it to the cloud database; Using a recursive character splitter to split the first score file into student document blocks, wherein each student document block includes all relevant subject scores and wrong question types in the student's exam; Obtain a dataset of students' historical performance and build a comprehensive evaluation model for data pair training to obtain individual performance summaries; Collect students' classroom performance images in real time, convert them into corresponding behavior labels, and upload them to the comprehensive evaluation model to obtain classroom seriousness coefficient scores; Combined with the individual performance summary and classroom performance data table, a correlation analysis is performed using a preset correlation analysis algorithm to obtain a comprehensive quality analysis; According to the input instructions, you can obtain the corresponding comprehensive quality analysis.

2. The classroom evaluation auxiliary method according to claim 1, characterized in that: The obtaining of the first score file is specifically as follows: The student's initial performance parameters include daily homework scores, test scores, and wrong question types; Normalize the collected student initial score parameters and set relevant score thresholds to remove invalid data; The first score file is set in a table format with student names as rows and subjects, wrong question types and scores as columns, and a timestamp is added to each row of data.

3. The classroom evaluation auxiliary method according to claim 1, characterized in that: The data is constructed to train a comprehensive evaluation model to obtain a summary of individual performance, specifically: The student historical performance data set includes historical student performance data and corresponding teacher comments; Preprocess the historical performance dataset, using the format of subject: score: wrong question type - teacher's comments as input to train a comprehensive evaluation model. Semantic analysis is used to perform part-of-speech tagging and extract keywords, and QLoRa is used to perform local fine-tuning on the comprehensive evaluation model; Get a summary of your personal performance through a comprehensive evaluation model.

4. The classroom evaluation auxiliary method according to claim 1, characterized in that: The method of obtaining the classroom seriousness coefficient score includes: Collect historical images related to the classroom and annotate relevant classroom behaviors; The supervised training method enables comprehensive evaluation of the large model's ability to acquire action recognition; The students' attitudes in class are obtained in real time during the period, and their class seriousness is quantified through the class seriousness coefficient.

5. The classroom evaluation auxiliary method according to claim 4, characterized in that: The calculation formula S of the classroom seriousness coefficient score is as follows: S = w1×A+w2×P+w3×D; Among them, A is the concentration, P is the classroom participation, D is the posture correction, and w1w2w3 are the corresponding weights respectively.

6. The classroom evaluation auxiliary method according to claim 1, characterized in that: The association analysis is performed using a preset association analysis algorithm to obtain a comprehensive quality analysis, specifically: Receive and merge the individual performance summary and class diligence coefficient scores output by the comprehensive evaluation model to obtain a comprehensive data set; The preset correlation analysis algorithm is used to correlate and analyze the comprehensive data set, and the confidence level is calculated to obtain a comprehensive quality analysis and upload it to the cloud database.

7. The classroom evaluation auxiliary method according to claim 1, characterized in that: The step of obtaining a corresponding comprehensive quality analysis according to the input instruction includes: Receive specific input instructions and identify key information in the instructions through keyword matching; According to the instruction analysis results, relevant comprehensive quality analysis is obtained from the cloud database.

8. A classroom evaluation auxiliary system, characterized in that: include: Score collection module, score analysis module, performance acquisition module and correlation analysis module; The score collection module collects students' initial score parameters, cleans the score parameters and enters them in a preset format to obtain a first score file, adds a timestamp to the first score file and uploads it to the cloud database; The performance analysis module uses a recursive character segmenter to segment the first performance file to obtain student document blocks, wherein each student document block includes scores of all relevant subjects and types of wrong questions in the student's examination; Obtain a student's historical performance data set, build a data pair and input it into a preset comprehensive evaluation model to obtain a summary of individual performance; The performance acquisition module collects students' classroom performance images in real time, converts them into corresponding behavior labels, and uploads them to the comprehensive evaluation model to obtain the classroom seriousness coefficient score; The association analysis module combines the individual performance summary and the classroom performance data table to perform association analysis using a preset association analysis algorithm to obtain a comprehensive quality analysis; According to the input instructions, you can obtain the corresponding comprehensive quality analysis.

9. A storage medium, characterized in that: The storage medium stores computer-readable program codes, which, when executed, implement the steps of a classroom evaluation assistance method according to any one of claims 1 to 7.

10. A terminal device, characterized in that: The method comprises a processor and a memory, wherein the memory stores a computer program, and when the processor executes the computer program, the steps of a classroom evaluation auxiliary method according to any one of claims 1 to 7 are implemented.

Images