College public course hybrid teaching control method and system

Abstract

The invention relates to the technical field of information control and internet platforms, and discloses a college public course hybrid teaching control method and system, and the method comprises the steps: a user terminal selects a teacher terminal or a student terminal through registration information, distributes different permissions according to the teacher terminal or the student terminal, and selects different modes; sending a student data collection instruction; obtaining students, carrying out statistical induction on the data, and carrying out index management and quality management on the student data after induction; performing student data prediction according to index management and quality management; analyzing a student data prediction result, endowing the student data prediction result with a classification label, and performing uplink processing on student data; performing an adjustment difficulty adjustment test on student data of the target student obtained in the block chain; and course pushing is carried out according to a test result. The system comprises a terminal module, a prediction module and a push module. According to the invention, by classifying the labels, predicting the student learning data and pushing the related data according to the predicted student learning data, the learning ability of the students and the student education management efficiency are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of information control and Internet platform, and in particular to a hybrid teaching control method and system for public courses in universities. Background Art

[0002] The hybrid teaching model is a combination of online and offline learning methods, with student participation and teacher-led activities to achieve the most ideal teaching results. One of the core differences between hybrid teaching and traditional teaching lies in the selection of teaching media and the design of teaching strategies. Traditional teaching media selection mainly considers how to better facilitate the teacher's presentation of teaching content; while the selection of teaching media in hybrid teaching focuses more on which media forms can best support student learning. Based on extensive research and practice, the implementation of hybrid teaching is technology-dependent, that is, hybrid teaching relies on a certain learning support platform. A mature learning platform is an important factor in ensuring the effectiveness of hybrid teaching.

[0003] Intelligent tutoring systems are becoming increasingly important in education. With the continuous advancement of artificial intelligence (AI), these systems can not only provide personalized learning experiences but also help students achieve better results in various subjects. By analyzing students' learning habits, knowledge acquisition, and progress, intelligent tutoring systems can develop study plans and exercises that are most suitable for them.

[0004] Prior Art 1, a Chinese patent application numbered 202411334266.X, discloses a long-sequence online learning prediction method for a smart education platform. This method relates to the field of artificial intelligence long-sequence prediction technology. The method involves collecting students' online learning status from an online education system to construct a multi-feature online learning status dataset. Sequences of a fixed number of historical time steps in the dataset are used as input to a model encoder and decoder. The original sequence X is fed into the multi-layer cascade encoder constructed by the model. The extracted encoded seasonal information features are output at the output, resulting in a seasonal information feature vector and a cyclical trend feature vector after encoding and decoding. The decoder output decomposition components are projected and fused to obtain a preliminary prediction sequence, and a portion of the preliminary prediction sequence is intercepted as the final prediction sequence. While this method can predict online learning status for long time-step sequences in the future and integrates multiple indicators related to student status to improve the reliability and accuracy of the prediction results, it suffers from poor platform compatibility and stability, impacting students' learning experience.

[0005] Prior art 2, a Chinese patent, application number: 202310444301.2, relates to the field of educational data prediction and processing technology, and in particular, to a smart campus education big data fusion method and platform. The method includes the following steps: acquiring student educational activity data; performing data standardization based on student educational activity data to obtain standard real-time data; fusing and constructing based on standard real-time data to construct a student behavior model; performing behavioral anomaly prediction based on the student behavior model to obtain abnormal behavior data; performing intelligent event tracing and evaluation based on abnormal behavior data to obtain detailed information on abnormal behavior for tracking abnormal events in smart campus education. Although the efficiency of student education management has been improved by acquiring, standardizing, and fusing student educational activity data, quickly constructing a student behavior model, and integrating multi-source data for comprehensive analysis, the online education platform is not well designed and fails to meet teaching needs.

[0006] Prior art three, Chinese patent, application number: 202211122849.7 relates to the field of data analysis technology, and in particular refers to a method and device for predicting a decline in learning quality. The method includes: receiving sleep information data of the object to be predicted; calculating a comprehensive sleep evaluation index of the object to be predicted based on the sleep information data; and performing a prediction based on the comprehensive sleep evaluation index using a pre-constructed learning quality decline prediction model to obtain the probability of a decline in learning quality of the object to be predicted. Although the relationship between sleep quality and learning quality can be quantitatively defined to a certain extent, and the probability of a decline in students' learning quality can be predicted, the prediction model lacks personalization and revealingness, cannot provide sufficient evidence to support the diagnosis and recommendation results, and is difficult to understand the specific basis for the prediction.

[0007] At present, the platform compatibility and stability of the existing technologies 1, 2 and 3 are poor, which affects the students' learning experience. The online education platform design is not perfect and fails to meet the teaching needs well. The prediction model lacks personalization and disclosure, cannot provide sufficient evidence to support the diagnosis and recommendation results, and is difficult to understand the specific basis of the prediction. The present invention provides a hybrid teaching control method and system for public courses in colleges and universities. Summary of the Invention

[0008] The main purpose of the present invention is to provide a hybrid teaching control method and system for public courses in colleges and universities, so as to solve the problems in the existing technology such as poor platform compatibility and stability, which affect students' learning experience, imperfect design of online education platforms, failure to meet teaching needs, lack of personalization and disclosure of prediction models, inability to provide sufficient evidence to support diagnosis and recommendation results, and difficulty in understanding the specific basis of prediction.

[0009] To achieve the above object, the present invention provides the following technical solutions:

[0010] A hybrid teaching control method for public courses in colleges and universities, comprising:

[0011] The user terminal selects the teacher terminal or the student terminal through registration information; different permissions are granted according to the teacher terminal or the student terminal; different modes are selected according to the different permissions; after the teacher terminal and the student terminal select the mode, a student data collection instruction is issued to collect student data;

[0012] Clean and convert the collected student data; perform statistical summarization on the converted data; perform indicator management and quality management on the summarized student data; and make student data predictions based on the indicator management and quality management;

[0013] Analyze student data prediction results, assign classification labels, and upload student data to the chain; adjust the difficulty of the test on the target student data obtained from the blockchain; and push courses based on the test results.

[0014] As a further improvement of the present invention, the process of issuing an instruction to collect student data includes:

[0015] The user terminal selects the teacher terminal or the student terminal by filling in information; the device receives the information submitted by the user terminal; and classifies the information into teacher terminal information and student terminal information according to the information classification rules;

[0016] According to the divided teacher-side information and student-side information, the corresponding account permissions are allocated in a mode; the teacher-side account corresponding to the information is allocated to online teaching or online and offline teaching mode, and the student-side account corresponding to the information is allocated to online learning or offline learning mode;

[0017] Among them, when the teacher chooses the online teaching mode, the student can only choose the online learning mode;

[0018] When the teacher and student sides have completed their selections, a collection instruction is issued to collect student learning data and teacher classroom teaching data, and the collected teacher classroom teaching data is stored in the teaching database.

[0019] As a further improvement of the present invention, the process of predicting student data includes:

[0020] Clean and transform the data collected in the teaching database, analyze and count the processed data to obtain a student data analysis table, and transfer the obtained student data analysis table to the knowledge database;

[0021] The knowledge database receives the student data analysis table, extracts the student data from the student data analysis table as knowledge information, and stores the knowledge information in the knowledge base; extracts the analyzed and statistically analyzed student data from the student analysis table as analysis information, and stores the analysis information in the analysis database;

[0022] Among them, knowledge information includes student-related knowledge and document data; analytical data includes homework completion details, extracurricular check-in details, test score details, and other personal data related to student learning;

[0023] Preset indicators, read the analysis information corresponding to the indicators from the analysis database, and form indicator statistical data based on the read analysis information; train students to learn prediction models based on the indicator statistical data.

[0024] As a further improvement of the present invention, the process of obtaining the student data analysis table includes:

[0025] Clean the data in the collected data database, delete outliers, and interpolate missing values; convert unstructured data into structured data and semi-structured data into structured data;

[0026] The conversion of unstructured data into structured data includes converting source data information in the teaching database into row data or field data and the corresponding relationship between the row data and field data, and writing the row data, field data and the corresponding relationship into the database;

[0027] Extract key features from the converted data and divide it according to the extracted key feature data; set division criteria and obtain different learning data types according to the division criteria; continuously optimize and adjust according to new data;

[0028] Among them, the classification criteria include students' academic performance, study habits, and participation in extracurricular activities;

[0029] Statistics are collected on different types of learning data to obtain a student data analysis table, which is then transferred to the knowledge database.

[0030] As a further improvement of the present invention, the process of storing the analysis information in the analysis database includes:

[0031] The student data analysis table extracts personal data related to student learning, including student learning knowledge, document data, homework completion details, extracurricular check-in details, and test score details;

[0032] Integrate the extracted data, clean the extracted and integrated data, remove invalid data, duplicate data and outliers; detect outliers by comparing data from different sources, and transform the cleaned data;

[0033] After cleaning and converting the data, knowledge information and analytical information are separated, the knowledge information is stored in the knowledge base, and the analytical information is stored in the analytical database.

[0034] As a further improvement of the present invention, the process of training students to learn the prediction model includes:

[0035] Preset indicators and read the analysis information corresponding to the indicators from the analysis database; generate indicator statistics based on the read analysis information; preset quality inspection standards, calibrate the indicator statistics, and eliminate indicator statistics with quality lower than the preset quality standard to obtain a standardized learning data set;

[0036] The standardized learning data set is divided into training set, validation set and test set in a ratio of 7:2:1 according to the indicator standard, and the student prediction model is constructed;

[0037] The student prediction model is trained according to the training set. The training set is input into the first block. The first block outputs the forward prediction result and the backward prediction result. The difference between the input data and the backward prediction result of the previous block is input into the next block, and then the forward prediction result and the backward prediction result of all blocks are obtained. The forward prediction result is integrated to obtain the stacked prediction result. The different stacked prediction results are added together to obtain the global prediction result; the final student learning prediction model is obtained.

[0038] As a further improvement of the present invention, the process of pushing courses according to the test results includes:

[0039] Input student learning data into the prediction model to obtain student data prediction results; construct multiple type labels, extract features from student data prediction results, classify the extracted features according to the type labels, and upload the type labels to the chain;

[0040] Adjust the difficulty test of student data based on the type label after being uploaded to the chain, simulate different learning scenarios and learning progress, evaluate students' learning effects at different levels of difficulty, set the evaluation effect threshold, and readjust if it is greater than the threshold; if it is less than the threshold, the difficulty is moderate;

[0041] Analyze the students' prediction results and the difficulty adjustment test results to obtain push keywords; retrieve courses from the knowledge base based on the push keywords, and push the retrieved courses from the knowledge base to students; the teacher receives the push keywords from the student side and answers them online.

[0042] As a further improvement of the present invention, the process of uploading the type tag to the chain includes:

[0043] Based on the predicted results of student data, we construct various types of labels and classify the predicted results of student data according to different learning status and learning effects; based on the classification results, we build a student learning management platform based on the blockchain network;

[0044] Integrate historical student learning data on the student learning management platform; store the student learning data integrated into the student learning management platform on the blockchain to form a trusted data record; compile smart contracts on the student learning management platform and configure the smart contracts to execute on the blockchain network;

[0045] Smart contracts are used to manage students’ learning process and progress, and to record students’ learning process and progress.

[0046] Build a student learning management platform based on the blockchain network, construct an intelligent development environment, set up a data interface for the student learning management platform, and integrate existing student learning data through the data interface;

[0047] The type tags received in real time are configured with uniquely coded identification information. The student learning management platform encrypts the student's learning status and learning results and stores them on the blockchain network. This completes on-chain storage and obtains a credible data record of the student's learning.

[0048] Obtain students' learning needs, write smart contracts in the intelligent development environment of the student learning management platform, and deploy the written smart contracts to the blockchain network;

[0049] Among them, students’ learning needs include improving learning efficiency, improving scores, and solving specific learning problems.

[0050] As a further improvement of the present invention, the process of evaluating students' learning effects at different levels of difficulty includes:

[0051] Obtain the preset weight corresponding to each preset dimension, and calculate a comprehensive score based on the target learning data information on each preset dimension and the corresponding preset weight;

[0052] The preset dimensions include learning time, answer accuracy, and learning progress; the target learning data information includes historical scores and current learning progress;

[0053] Calculate the mean of the initial learning data on each preset dimension as the target learning data information; compare the target learning data with the preset standard value to determine the difficulty information of the preset dimension; determine the target difficulty based on the difficulty information, and adjust the test difficulty based on the target difficulty;

[0054] Simulate different learning scenarios and learning accuracy, and evaluate students' learning effects at different difficulty levels through simulated scenarios; set an evaluation effect threshold. If the evaluation effect is greater than the threshold, the current test difficulty is too high and needs to be readjusted; if it is less than the threshold, the test difficulty is moderate.

[0055] To achieve the above object, the present invention also provides the following technical solutions:

[0056] A hybrid teaching control system for public courses in universities, comprising:

[0057] The terminal module is responsible for enabling the user terminal to select the teacher terminal or the student terminal through registration information; granting different permissions based on the teacher terminal or the student terminal; and selecting different modes based on the permissions. After the teacher terminal and the student terminal have completed the mode selection, they issue a student data collection instruction to collect student data; and output the collected student data to the prediction module.

[0058] Among them, permissions include teacher side and student side; mode includes teaching method or learning method; teacher side permission allocation selects online teaching or online and offline teaching mode, while student side permission allocation selects online learning or offline learning mode;

[0059] The prediction module is responsible for cleaning and converting the collected student data; performing statistical summarization on the converted data; performing indicator management and quality management on the summarized student data; predicting student data based on the indicator management and quality management; and outputting the prediction results to the push module;

[0060] The push module is responsible for analyzing student data prediction results, assigning classification labels, and uploading student data to the chain; adjusting the difficulty of the student data of the target students obtained from the blockchain; and pushing courses based on the test results.

[0061] The user terminal of the present invention selects the teacher end or the student end through registration information, and different permissions are granted according to the selected port. The teacher end can teach online or online and offline, and the student end can choose online learning or offline learning mode; according to different permissions, users can choose different teaching or learning methods to personalize the teaching and learning experience; after the teacher end and the student end complete the mode selection, the system issues an instruction to collect student data, and collects student data for analysis and prediction; the collected student data is cleaned and converted for statistical analysis and model training; the converted data is statistically summarized to extract information and features; according to the results of the statistical induction, indicator management and quality management are carried out, and according to the results of the indicator management and quality management, student data is predicted to timely discover possible problems of students and take corresponding measures; the student data prediction structure is analyzed and classification labels are assigned to help identify students' different learning states and needs; the student data is processed on the chain to increase the security and immutability of the data and enhance the credibility of the data; the student data of the target students obtained in the blockchain is adjusted for difficulty testing to match the teaching content with the student's ability; away push is carried out according to the test results, and personalized learning paths and resources are provided to improve learning effects. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] Figure 1 This is a schematic diagram of the steps of an embodiment of the hybrid teaching control method for public courses in colleges and universities of the present invention;

[0063] Figure 2 This is a flowchart of the steps of issuing a student data collection instruction according to an embodiment of the hybrid teaching control method for public courses in universities of the present invention;

[0064] Figure 3 This is a flow chart of the steps of predicting student data in one embodiment of the hybrid teaching control method for public courses in universities of the present invention;

[0065] Figure 4 This is a flow chart of the steps of obtaining a student data analysis table according to an embodiment of the hybrid teaching control method for public courses in universities of the present invention;

[0066] Figure 5 This is a flow chart of the steps of storing analysis information into an analysis database in one embodiment of the hybrid teaching control method for public courses in universities of the present invention;

[0067] Figure 6 This is a flowchart of the steps of training a student learning prediction model according to an embodiment of the hybrid teaching control method for public courses in colleges and universities of the present invention;

[0068] Figure 7This is a flow chart of the steps of pushing courses according to test results in one embodiment of the hybrid teaching control method for public courses in universities of the present invention;

[0069] Figure 8 This is a flow chart of the steps for uploading type tags to a chain in accordance with an embodiment of the hybrid teaching control method for public courses in universities of the present invention;

[0070] Figure 9 This is a flowchart of the steps for configuring a smart contract to a blockchain network according to an embodiment of the hybrid teaching control method for public courses in universities of the present invention;

[0071] Figure 10 This is a flowchart of steps for evaluating students' learning effects at different levels of difficulty in one embodiment of a hybrid teaching control method for public courses in universities of the present invention;

[0072] Figure 11 This is a functional module diagram of an embodiment of the hybrid teaching control system for public courses in universities of the present invention;

[0073] Figure 12 This is a schematic structural diagram of an embodiment of an electronic device of the present invention;

[0074] Figure 13 This is a schematic structural diagram of an embodiment of a storage medium of the present invention. DETAILED DESCRIPTION

[0075] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0076] The terms "first", "second" and "third" in the present invention are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first", "second" and "third" may explicitly or implicitly include at least one of the features. In the description of the present invention, the meaning of "multiple" is at least two, for example, two, three, etc., unless otherwise clearly and specifically defined. All directional indications in the embodiments of the present invention (such as up, down, left, right, front, back...) are only used to explain the relative position relationship, movement, etc. between the components under a certain specific posture (as shown in the accompanying drawings). If the specific posture changes, the directional indication will also change accordingly. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units that are not listed, or may optionally include other steps or units that are inherent to these processes, methods, products or devices.

[0077] References herein to "embodiments" mean that a particular feature, structure, or characteristic described in connection with the embodiments may be included in at least one embodiment of the present invention. The appearance of this phrase in various places in the specification does not necessarily refer to the same embodiment, nor does it constitute a separate or alternative embodiment that is mutually exclusive of other embodiments. It is understood, both explicitly and implicitly, by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0078] like Figure 1 As shown, this embodiment provides an embodiment of a hybrid teaching control method for public courses in colleges and universities. In this embodiment, the hybrid teaching control method for public courses in colleges and universities specifically includes the following steps:

[0079] Step S1: The user terminal selects the teacher terminal or the student terminal through registration information; different permissions are granted according to the teacher terminal or the student terminal; different modes are selected according to the different permissions; after the teacher terminal and the student terminal select the mode, a student data collection instruction is issued to collect student data;

[0080] Among them, permissions include teacher side and student side; mode includes teaching method or learning method; teacher side permission allocation selects online teaching or online and offline teaching mode, while student side permission allocation selects online learning or offline learning mode;

[0081] Step S2: Clean and convert the collected student data; perform statistical summarization on the converted data; perform index management and quality management on the summarized student data; and perform student data prediction based on the index management and quality management;

[0082] Step S3: Analyze the student data prediction results, assign classification labels, and upload the student data to the blockchain; adjust the difficulty of the student data of the target students obtained from the blockchain; and push courses based on the test results.

[0083] Preferably, in step S1 of this embodiment, the user terminal selects the teacher terminal or the student terminal through registration information, and is granted different permissions according to the selected port. The teacher terminal can teach online or online and offline, and the student terminal can choose online learning or offline learning mode; according to different permissions, users can choose different teaching or learning methods to personalize the teaching and learning experience; after the teacher terminal and the student terminal select the mode, the system issues an instruction to collect student data, and collects student data for analysis and prediction; in step S2, the collected student data is cleaned and the cleaned data is converted for statistical analysis and model training; the converted data is statistically summarized to provide The system obtains information and features; conducts indicator management and quality management based on the results of statistical induction, and conducts student data prediction based on the results of indicator management and quality management, timely discovers possible problems of students, and takes corresponding measures; in step S3, the student data prediction structure is analyzed and classification labels are assigned to help identify students' different learning status and needs; student data is processed on the chain to increase the security and immutability of the data and enhance the credibility of the data; the student data of the target students obtained from the blockchain are adjusted for difficulty testing to match the teaching content with the students' abilities; away game push is carried out based on the test results, and personalized learning paths and resources are provided to improve learning effects.

[0084] Therefore, this embodiment implements full-process management from data collection to personalized teaching, using blockchain technology to ensure data security and credibility. It also improves data quality through data cleaning and conversion, enabling precise teaching and enhanced learning outcomes. It also establishes an adaptive evaluation and feedback loop mechanism. Adaptive test design can adjust the difficulty of tests based on the learner's ability level to ensure the accuracy and fairness of the evaluation results. The system should be able to evaluate learners' learning outcomes in real time based on their performance and provide targeted feedback and suggestions. At the same time, learners can also provide feedback to the system through the feedback system to improve their personalized learning plans, thereby promoting the continuous improvement and optimization of the adaptive system.

[0085] Further, if Figure 2 As shown, the process of issuing the instruction to collect student data in step S1 specifically includes the following steps:

[0086] Step S11: The user terminal selects the teacher terminal or the student terminal by filling in information; the device receives the information submitted by the user terminal and classifies the information into teacher terminal information and student terminal information according to the information classification rules;

[0087] Step S12: assigning permissions to corresponding accounts based on the divided teacher-side information and student-side information; assigning online teaching or online and offline teaching mode to the teacher-side account corresponding to the information, and assigning online learning or offline learning mode to the student-side account corresponding to the information;

[0088] Among them, when the teacher chooses the online teaching mode, the student can only choose the online learning mode;

[0089] The permission value is calculated by combining multiple factors and weights to ensure the accuracy of permission allocation:

[0090]

[0091] In the formula, P represents the permission value, which indicates the final assigned permission level. A represents the activity factor, which indicates the current activity status of the teacher or student, such as online status and activity level. The higher the activity factor, the greater the permission value. T represents the teaching mode factor, which indicates the teaching mode selected by the teacher (such as online teaching, offline teaching, hybrid teaching, etc.). Different teaching modes correspond to different factor values. S represents the student engagement factor, which indicates student participation and learning enthusiasm. The higher the engagement, the larger the factor value. M represents the management factor, which indicates the degree of intervention of the system administrator in permission allocation. The higher the management factor, the stricter the system control over permissions. W represents the weight factor, which indicates the weight of different factors in the calculation. The weight factor can be adjusted according to actual conditions. C represents the course difficulty factor, which indicates the difficulty level of the course. The higher the difficulty, the larger the factor value. D represents the data integrity factor, which indicates the integrity and accuracy of the data. The more complete the data, the larger the factor value. The activity factor A indicates that the activity factor reflects the online status and activity level of the teacher or student. For example, if the teacher or student is frequently online and active, the value of A will be higher, thereby increasing the permission value P. The teaching mode factor T indicates that the teaching mode factor adjusts the permission value according to the teaching mode selected by the teacher. For example, an online teaching mode may have a lower factor value, while a hybrid teaching mode may have a higher factor value; the student participation factor S indicates that the student participation factor reflects the student's enthusiasm for learning. If students actively participate in class activities, the value of S will be higher, thereby increasing the permission value P; the management factor M indicates that the management factor allows the system administrator to adjust the strictness of permission allocation as needed. For example, in some special cases, the administrator may want to increase control over permissions, in which case the value of M will be higher; the weight factor W indicates that the weight factor is used to adjust the importance of different factors in the calculation. For example, if a certain factor has a greater impact on permission allocation, its weight factor can be increased; the course difficulty factor C indicates that the course difficulty factor reflects the difficulty level of the course. The higher the difficulty, the larger the factor value, thereby increasing the permission value P; the data integrity factor D indicates that the data integrity factor reflects the integrity and accuracy of the data. The more complete the data, the larger the factor value, thereby increasing the permission value P;

[0092] Step S13: After the teacher and student terminals have completed their selections, a collection instruction is issued to collect student learning data; and teacher classroom teaching data is collected, and the collected teacher classroom teaching data is stored in a teaching database.

[0093] Preferably, in step S11 of this embodiment, the user selects a role by filling in information, the device receives and processes the information, and the user can select the corresponding role according to his or her needs to enter the appropriate system interface, thereby improving the user experience and laying the foundation for data processing and authority allocation; step S12 involves authority management and data analysis, and different authorities and teaching modes are allocated according to user roles; through reasonable rights allocation, teaching activities are made more standardized and secure; at the same time, the student end is restricted to intelligently select online learning when the teacher end selects the online teaching mode, to ensure the consistency of the teaching mode; in step S13, after issuing a collection instruction, the student's learning data and the teacher's teaching data are collected and stored in the database; through real-time data collection and storage, basic data support is provided for teaching analysis and optimization, and teaching strategies are adjusted according to the data.

[0094] Therefore, this embodiment constitutes a complete data processing flow, from user interaction to data classification, authority allocation, and data collection and storage, each step provides technical support for improving teaching quality and management efficiency.

[0095] Further, if Figure 3 As shown, the process of predicting student data in step S2 specifically includes the following steps:

[0096] Step S21: Cleaning and converting the data collected in the teaching database, analyzing and statistically analyzing the processed data to obtain a student data analysis table, and transferring the obtained student data analysis table to the knowledge database;

[0097] Step S22: The knowledge database receives the student data analysis table, extracts student data from the student data analysis table as knowledge information, and stores the knowledge information in the knowledge base; extracts the analyzed and statistically analyzed student data from the student analysis table as analysis information, and stores the analysis information in the analysis database;

[0098] Among them, knowledge information includes student-related knowledge and document data; analytical data includes homework completion details, extracurricular check-in details, test score details, and other personal data related to student learning;

[0099] Step S23: Preset indicators, read analysis information corresponding to the indicators from the analysis database, and form indicator statistical data based on the read analysis information; train students to learn prediction models based on the indicator statistical data.

[0100] Preferably, in step S21 of this embodiment, the accuracy and consistency of the data are ensured by identifying and processing outliers, missing values and noise data; the original data is converted into a format suitable for analysis, such as standardization, normalization, etc., to facilitate subsequent statistical analysis. The processed data is summarized and counted to generate a student data analysis table to provide basic data for the knowledge base and analysis database; the generated student data analysis table is transferred to the knowledge database to facilitate the extraction of knowledge information and the storage of analysis information. In step S22, knowledge information related to student learning, such as learning content, file data, etc., is extracted from the student data analysis table and stored in the knowledge base; student data that has been analyzed and counted, such as homework completion status, test scores, etc., is extracted from the student data analysis table and stored in the analysis database; the knowledge information and analysis information are stored in the knowledge base and analysis database respectively to facilitate subsequent query and analysis. In step S23, analysis indicators are preset according to educational needs, such as homework completion rate, test scores, etc., to provide direction for subsequent data analysis; analysis information corresponding to the indicators is read from the analysis database, and statistical analysis is performed to form indicator statistical data; a student learning prediction model is trained based on the indicator statistical data to predict students' learning performance and potential problems. Through data analysis and model training, the scientificity and accuracy of educational decision-making are improved, helping teachers to better understand students' learning situation and optimize teaching strategies.

[0101] Therefore, this embodiment improves the efficiency and quality of data processing, and also provides strong technical support for educational informatization and personalized learning through data analysis and prediction model training.

[0102] Furthermore, if Figure 4 As shown, the process of obtaining the student data analysis table in step S21 specifically includes the following steps:

[0103] Step S211: Clean the data in the collected data database, delete outliers, and interpolate missing values; convert unstructured data into structured data and convert semi-structured data into structured data;

[0104] The conversion of unstructured data into structured data includes converting source data information in the teaching database into row data or field data and the corresponding relationship between the row data and field data, and writing the row data, field data and the corresponding relationship into the database;

[0105] Step S212: extract key features from the converted data and divide them according to the extracted key feature data; set division criteria and obtain different learning data types according to the division criteria; and continuously optimize and adjust according to new data;

[0106] Among them, the classification criteria include students' academic performance, study habits, and participation in extracurricular activities;

[0107] Step S213: Perform statistics on different types of learning data to obtain a student data analysis table, and transmit the student data analysis table to the knowledge database.

[0108] Preferably, in step S211 of this embodiment, outliers are deleted and missing values are interpolated. Outlier handling strategies include deletion, correction, and replacement. Missing value handling strategies include deletion, filling, or retention strategies; unstructured data (such as text, audio, and video) is converted into structured data (such as row data and field data). By deleting outliers and interpolating missing values, the accuracy and completeness of the data are ensured, thereby improving the reliability of subsequent analysis. After converting unstructured data into structured data, data analysis and modeling can be more convenient, improving the efficiency and effectiveness of data processing. In step S212, key features are extracted from the converted data. These features may include student grades, study habits, and participation in extracurricular activities. Data is partitioned based on the extracted key features, and partitioning criteria are set to obtain different learning data types. Through key feature extraction and data partitioning, the data structure is optimized, making the data more suitable for specific learning analysis tasks. Different learning data types can be used to support personalized learning analysis, help identify students' learning needs, and improve teaching strategies. In step S213, statistics are collected on the different learning data types to generate a student data analysis table. The generated student data analysis table is transferred to the knowledge database for further analysis and application. This data analysis table supports educational institutions and teachers in decision-making, such as personalized teaching plans and resource allocation. By analyzing student learning data, learning difficulties and areas for improvement can be identified, thereby improving overall learning outcomes.

[0109] Therefore, this embodiment ensures the quality and availability of data through technical means such as data cleaning, feature extraction, data partitioning and statistical analysis, providing a solid foundation for learning analysis and decision support.

[0110] Further, if Figure 5 As shown, the process of storing the analysis information into the analysis database in step S22 specifically includes the following steps:

[0111] Step S221: Extracting personal data related to student learning, such as student learning-related knowledge, document data, homework completion details, extracurricular check-in details, and test score details, from the student data analysis table;

[0112] Step S222: Integrate the extracted data, clean the extracted and integrated data to remove invalid data, duplicate data, and outliers; detect outliers by comparing data from different sources, and transform the cleaned data;

[0113] Step S223: After cleaning and converting the data, separate the knowledge information and the analysis information, store the knowledge information in the knowledge base, and store the analysis information in the analysis database.

[0114] Preferably, step S221 of this embodiment extracts data related to student learning from multiple data sources, including information such as learning knowledge, document data, homework completion status, extracurricular check-ins, and test scores. By extracting this data, a comprehensive student learning dataset can be constructed, providing foundational support for subsequent data analysis and decision-making. Understanding students' learning behaviors and academic performance can provide a basis for educational improvement. Step S222 involves data integration, cleaning, and conversion. First, data from different sources is integrated into a unified dataset. Then, data cleaning is performed to remove invalid data, duplicate data, and outliers. The cleaning process utilizes various techniques and algorithms to identify and correct data quality issues. Furthermore, outliers are detected by comparing data from different sources, and the cleaned data undergoes format and type conversion. Data integration and cleaning ensure data consistency and accuracy, improving the reliability of data analysis. This helps reduce erroneous analysis results caused by data quality issues, thereby improving the quality of decision support. Step S223 involves separating the cleaned and converted data into knowledge information and analytical information. Knowledge information is typically used for long-term storage and querying, while analytical information is used for immediate analysis and report generation. This information is stored in separate databases, such as a knowledge base and an analytical database. Separating and storing different types of information improves data management efficiency and flexibility. The knowledge base allows for long-term knowledge accumulation and querying, while the analytical database facilitates rapid data access and analysis, supporting real-time decision-making.

[0115] Therefore, this embodiment ensures the quality and availability of data through data extraction, integration, cleaning, conversion and separation, providing a solid foundation for subsequent education data analysis and decision-making.

[0116] Further, if Figure 6 As shown, the process of training students to learn the prediction model in step S23 specifically includes the following steps:

[0117] Step S231: Preset an indicator and read analysis information corresponding to the indicator from the analysis database; generate indicator statistical data based on the read analysis information; preset a quality inspection standard, calibrate the indicator statistical data, and eliminate indicator statistical data with quality lower than the preset quality standard to obtain a standardized learning data set;

[0118] Step S232: Divide the standardized learning data set into a training set, a validation set, and a test set in a ratio of 7:2:1 according to the indicator standard, and construct a student prediction model;

[0119] Step S233: Train the student prediction model according to the training set, input the training set into the first block, the first block outputs the forward prediction result and the backward prediction result, input the difference between the input data of the previous block and the backward prediction result into the next block, and then obtain the forward prediction result and the backward prediction result of all blocks, integrate the forward prediction result to obtain the stacked prediction result, add up the different stacked prediction results to obtain the global prediction result; and obtain the final student learning prediction model.

[0120] Preferably, step S231 of this embodiment presets indicators, reads analysis information corresponding to the indicators from the analysis database; forms indicator statistics based on the read analysis information; presets quality inspection standards, calibrates the indicator statistics, eliminates indicator statistics with quality lower than the preset quality standards, and obtains a standardized learning data set. By presetting indicators and quality inspection standards, the accuracy and reliability of the data set are guaranteed, providing a high-quality data foundation for subsequent model training. Step S232 divides the standardized learning data set into a training set, a validation set, and a test set in a ratio of 7:2:1 to construct a student prediction model. The division ratio helps to effectively tune and verify the model during the training process, and at the same time, the final performance of the model is evaluated through the test set to ensure the generalization ability of the model. Step S233 trains the student prediction model based on the training set. The training set is input into the first block, which outputs forward and backward prediction results. The difference between the input data and the backward prediction result of the previous block is input into the next block, thereby obtaining the forward and backward prediction results of all blocks. The forward prediction results are integrated to obtain a stacked prediction result, and the different stacked prediction results are added together to obtain a global prediction result, thereby obtaining the final student learning prediction model. By stacking and integrating the prediction results at multiple levels, the accuracy and stability of the prediction are improved, and the final global prediction result can more comprehensively reflect the student's learning situation.

[0121] Therefore, this embodiment ensures the accuracy and reliability of the student learning prediction model through technical means such as preprocessing, data partitioning, model training and prediction integration, providing strong support for education management and decision-making.

[0122] Furthermore, the process of obtaining the global prediction result in step S233 specifically includes the following steps:

[0123] Step S2331: During the training of the student learning prediction model, each module (block) outputs a positive prediction result. The prediction result reflects the student learning prediction model's understanding and prediction ability of student learning behavior at different levels. Different weights are assigned to each module based on its importance in the model and its prediction accuracy. For example, the prediction results of the bottom-level module may focus more on the processing of basic data, while the prediction results of the high-level module may focus more on the extraction of complex features.

[0124] Multiply the forward prediction result of each module by its corresponding weight, and then perform weighted summation; introduce dynamic adjustment to dynamically adjust the weight of each module based on the performance of the validation set;

[0125] Step S2332: finally generating a stacked prediction result through a hierarchical integration strategy; performing weighted fusion on different stacked prediction results, assigning weights based on the accuracy and stability of each stacked prediction result; multiplying each stacked prediction result by its corresponding weight, and performing a global weighted summation;

[0126] Step S2333: During the fusion process, a dynamic optimization mechanism is introduced to dynamically adjust the weights of each stacked prediction result based on the performance of the test set; a global prediction result is finally generated through the fusion strategy of multiple stacked prediction results;

[0127] The process of dynamically adjusting the weights of each stacked prediction result specifically includes:

[0128] For each stacked prediction result, initialize a weight value; use random initialization or set the initial value based on experience;

[0129] Assuming there are three stacked prediction results, the initial weights can be set to [w1,w2,w3]=[0.33,0.33,0.34];

[0130] Calculate the gradient. In each iteration, you need to calculate the gradient of each stacked prediction result. The gradient represents the sensitivity of the weight to the loss function, that is, the impact of the weight change on the loss function. Define a loss function to measure the gap between the prediction result and the true value. Use the mean square error (MSE) as the loss function:

[0131]

[0132] Among them, y i is the true value, is the predicted value, N is the number of samples;

[0133] For each stacked prediction result, the weight w i , calculate its gradient with respect to the loss function

[0134]

[0135] The specific calculation method depends on the form of the loss function. For example, for the mean square error loss function, the gradient is calculated as follows:

[0136]

[0137] Use an adaptive optimization algorithm to dynamically adjust the weight of each stacked prediction result based on the calculated gradient;

[0138] Adam optimization algorithm:

[0139] First moment estimate (mean of gradient):

[0140]

[0141] Among them, m t is the first-order moment estimate of the t-th iteration, m t-1 is the first-order moment estimate of the t-1th iteration, β1 is the exponential decay rate;

[0142] Second-order moment estimate** (mean of the square of the gradient):

[0143]

[0144] Among them, `v_t` is the second-order moment estimate of the `t`th iteration, v t-1 is the second-order moment estimate at the t-1th iteration, and β2 is the exponential decay rate.

[0145] Bias correction:

[0146]

[0147] in, and are the bias-corrected estimates of the first and second moments.

[0148] Weight update:

[0149]

[0150] Iterative Optimization

[0151] Repeat the above steps until the preset number of iterations is reached or the loss function converges. In each iteration, the gradient of each stacked prediction result is calculated and the weights are updated using an adaptive optimization algorithm. Based on the performance of the validation set, the learning rate and optimizer parameters are dynamically adjusted to optimize the weight update process. When the change in the loss function is less than a certain threshold or the preset number of iterations is reached, the model is considered to have converged and the iteration is stopped. After the weight update is completed, the final weight is used to perform a weighted summation of the stacked prediction results to generate a global prediction result.

[0152] Assume there are three stacked prediction results [y1,y2,y3], the final global prediction result y global The calculation is as follows:

[0153] y global =w1·y1+w2·y2+w3·y3

[0154] Through the above detailed steps, the weight update speed can be dynamically adjusted according to the size of the gradient, thereby improving the convergence speed and stability of the model during training;

[0155] Preferably, step S2331 of this embodiment performs a weighted summation of the module forward prediction results. The forward prediction results of each module reflect the model's understanding and predictive ability of students' learning behaviors at different levels. Low-level modules may focus on processing basic data, while high-level modules may focus on extracting complex features. Different weights are assigned to each module based on its importance in the model and its prediction accuracy. This ensures that during the weighted summation process, more important modules have a greater impact on the final result. A dynamic adjustment mechanism is introduced to dynamically adjust the weights of each module based on the performance of the validation set, enabling the model to adaptively adjust its structure to better adapt to different data distributions and learning behaviors. Significance: Through multi-level understanding and prediction, as well as dynamic adjustment of weights, the model can more accurately capture students' learning behaviors, thereby improving prediction accuracy. The dynamic adjustment mechanism enables the model to adapt to different data environments and learning behaviors, enhancing the model's robustness and generalization capabilities. Step S2332 performs a weighted fusion of hierarchical integration and stacked prediction results. Through a hierarchical integration strategy, the prediction results of different levels are integrated to generate a stacked prediction result. This helps to integrate prediction information at different levels and improve the comprehensiveness of the prediction; different stacked prediction results are weighted and fused, and weights are assigned according to the accuracy and stability of each stacked prediction result. This ensures that during the fusion process, more accurate and stable results have a greater impact on the final prediction result; multiplying each stacked prediction result by its corresponding weight and performing a global weighted sum helps to generate a comprehensive global prediction result. Significance: Through hierarchical integration and weighted fusion, the model can integrate prediction information at different levels and generate more comprehensive and accurate prediction results; through weighted fusion, the model can balance the accuracy and stability of different prediction results, thereby enhancing the stability of the prediction results. Step S2333 dynamically optimizes the fusion of multiple stacked prediction results and dynamically adjusts the weights of each stacked prediction result based on the performance of the test set; enables the model to adaptively adjust its structure to better adapt to different data distributions and learning behaviors; through the fusion strategy of multiple stacked prediction results, ultimately generates a global prediction result; helps to integrate information from different stacked prediction results and improve the accuracy and stability of the prediction. Significance: The dynamic optimization mechanism enables the model to adapt to different data environments and learning behaviors, improving the adaptability and generalization ability of predictions. By fusing multiple stacked prediction results, the model can integrate information from different prediction results to generate more accurate and stable predictions.

[0156] In summary, the global prediction results generated in this embodiment not only improve the accuracy and stability of the predictions, but also enhance the robustness and adaptability of the model. This enables the model to better understand and predict students' learning behaviors, providing strong support for educational decision-making and personalized learning.

[0157] Furthermore, if Figure 7 As shown, the process of pushing courses according to the test results in step S3 specifically includes the following steps:

[0158] Step S31: Input student learning data into the prediction model to obtain student data prediction results; construct multiple type labels, extract features from the student data prediction results, classify the feature extraction according to the type labels, and upload the type labels to the chain;

[0159] Step S32: Adjust the difficulty of the student data according to the type label after being uploaded to the chain, simulate different learning scenarios and learning progress, evaluate the student's learning effect at different levels of difficulty, set the evaluation effect threshold, and readjust if it is greater than the threshold; if it is less than the threshold, the difficulty is moderate;

[0160] Step S33: Analyze the student prediction results and the difficulty adjustment test results to obtain push keywords; retrieve courses in the knowledge base based on the push keywords, and push the retrieved courses in the knowledge base to students; the teacher receives the push keywords from the student and answers them online.

[0161] Preferably, in step S31 of this embodiment, student learning data is input into a prediction model to generate prediction results. Multiple type labels are then constructed to extract features and classify the student data. This utilizes feature extraction techniques from machine learning, selecting feature vectors through normalization, mean vector calculation, and other methods for mapping, thereby optimizing classification results. Furthermore, the on-chain processing involves blockchain technology to ensure data security and immutability. In step S32, a difficulty adjustment test is performed on the student data based on the on-chain type labels, simulating different learning scenarios and progress levels to evaluate student learning outcomes at varying levels of difficulty. This involves question difficulty assessment techniques from educational data mining, setting assessment thresholds to determine whether the difficulty is appropriate. This adjustment method facilitates personalized teaching and improves student learning efficiency. In step S33, the student prediction results and the difficulty adjustment test results are analyzed to derive push keywords, and course resources are retrieved from the knowledge base based on these keywords. The teacher receives the push keywords from the student and provides online answers, enhancing interactive and timely teaching.

[0162] Therefore, this embodiment improves the ability to accurately predict students' learning status, and can effectively analyze students' learning behavior and knowledge mastery through deep learning methods. Through feature extraction and classification processing, students' learning needs can be understood more carefully, thereby providing personalized learning support. On-chain processing may involve data security and traceability, which helps to protect student privacy and ensure data integrity. By simulating learning scenarios of different difficulty levels, students' learning effects can be evaluated more accurately, helping teachers and systems to better adjust teaching strategies. Setting evaluation effect thresholds helps to ensure the rationality of learning difficulty and avoid adverse effects on students from overly challenging or overly simple content. By analyzing prediction results and test results, personalized learning resources and path recommendations can be provided to students. The real-time interactive answer function on the teacher side enhances students' sense of participation and learning experience, and helps to solve students' learning problems in a timely manner.

[0163] Furthermore, if Figure 8 As shown, the process of uploading the type tag to the chain in step S31 specifically includes the following steps:

[0164] Step S311: Based on the student data prediction results, construct multiple types of tags, classify the student data prediction results according to different learning status and learning effects; and build a student learning management platform based on the blockchain network based on the classification results;

[0165] Step S312: Integrate historical student learning data on the student learning management platform; store the student learning data integrated into the student learning management platform on the blockchain to form a trusted data record; compile a smart contract on the student learning management platform and configure the smart contract to execute on the blockchain network;

[0166] Step S313: Manage the student’s learning process and progress through smart contracts, and record the student’s learning process and progress.

[0167] Preferably, in step S311 of this embodiment, multiple types of labels are constructed using the student data prediction results and categorized according to different learning statuses and learning outcomes. This provides a more accurate understanding of students' learning status, providing a basis for subsequent educational management. By categorizing the student data prediction results, dynamic monitoring of students' learning status and personalized guidance can be achieved, improving the relevance and effectiveness of education. In step S312, historical student learning data is integrated using blockchain technology and stored on-chain, forming a trusted data record. The decentralization, security, transparency, and immutability of blockchain technology ensure the authenticity and reliability of the data. Furthermore, through the development and execution of smart contracts, data can be automatically managed and updated, thereby improving the efficiency and accuracy of data processing. In step S313, smart contracts are used to manage and record students' learning process and progress. The automatic execution of smart contracts enables real-time monitoring and recording of various operations during the learning process, ensuring the transparency and traceability of learning activities. This mechanism not only helps improve the efficiency of learning management but also provides students with real-time feedback, helping them better adjust their learning strategies.

[0168] Therefore, this embodiment not only improves the efficiency and transparency of education management, but also enhances the security and credibility of data, providing students with a more personalized and high-quality learning experience.

[0169] Furthermore, if Figure 9 As shown, the process of configuring the smart contract to be executed on the blockchain network in step S312 specifically includes the following steps:

[0170] Step S3121: Build a student learning management platform based on the blockchain network, construct an intelligent development environment, set up a data interface for the student learning management platform, and integrate existing student learning data through the data interface;

[0171] Step S3122: The type tag received in real time is configured with uniquely coded identification information. The student learning management platform encrypts the student's learning status and learning results and stores them on the blockchain network. This completes on-chain storage and obtains a credible data record of the student's learning.

[0172] Step S3123: Obtain student learning needs, write a smart contract in the smart development environment of the student learning management platform, and deploy the written smart contract to the blockchain network;

[0173] Among them, students’ learning needs include improving learning efficiency, improving scores, and solving specific learning problems.

[0174] Preferably, in step S3121 of this embodiment, a student learning management platform based on a blockchain network is established, along with an intelligent development environment. By setting up a data interface, existing student learning data can be integrated, enabling real-time data sharing and collaboration. This improves data accessibility and integration, enabling student learning data to be managed and analyzed on a unified platform. In step S3122, uniquely coded identification information is assigned to the type tags received in real time, and the student's learning status and results are encrypted and stored on the blockchain network. This completes on-chain storage, resulting in a reliable data record. Blockchain technology ensures data security and immutability, thereby generating a reliable learning record. This not only improves data security but also enhances data transparency and credibility, allowing parents, teachers, and students to review learning progress at any time, enabling home-school communication. In step S3123, student learning needs are captured, and smart contracts are written in the intelligent development environment and deployed on the blockchain network. Smart contracts are used to automatically address student learning needs, such as improving learning efficiency and increasing grades. The deployment and execution of smart contracts rely on blockchain platforms, such as the Ethereum Virtual Machine. The effect of this technology is to simplify the learning process, reduce costs and improve efficiency by automating contracts while ensuring the correctness and security of the contracts.

[0175] Therefore, this embodiment enables the blockchain-based student learning management platform to be efficient, secure, transparent and intelligent, thereby providing students with a more reliable and personalized learning environment.

[0176] Further, if Figure 10 As shown, the process of evaluating students' learning effects at different levels of difficulty in step S32 specifically includes the following steps:

[0177] Step S321: Obtain the preset weight corresponding to each preset dimension, and calculate a comprehensive score based on the target learning data information on each preset dimension and the corresponding preset weight;

[0178] The preset dimensions include learning time, answer accuracy, and learning progress; the target learning data information includes historical scores and current learning progress;

[0179] The calculation expression of the comprehensive score is:

[0180]

[0181] In the formula, U represents the comprehensive score =, which represents the comprehensive evaluation of students' learning effects at different difficulty levels, L represents the learning time, which represents the learning time of students on a certain preset dimension, B represents the correct answer rate, which represents the correct answer rate of students on a certain preset dimension, Q represents the learning progress, which represents the learning progress of students on a certain preset dimension, and WL Indicates the learning time weight, which indicates the weight of learning time in the calculation of comprehensive score, W B Indicates the weight of the correct answer rate, which indicates the weight of the correct answer rate in the calculation of the comprehensive score, W Q Indicates the learning progress weight, which indicates the weight of learning progress in the calculation of comprehensive scores. N represents the normalization factor, which is used to normalize the weights of each dimension, usually the sum of the weights. D T Indicates the target difficulty, which is determined based on the target learning data and the preset standard value. C Indicates the current difficulty, indicating the difficulty information of the current test, E U It represents learning efficiency, which indicates the efficiency of students in the learning process. It is usually calculated by learning time and learning progress. Q Indicates learning motivation, which indicates the student's motivation in the learning process, usually calculated by learning time and correct answer rate;

[0182] Step S322: Calculate the mean of the initial learning data on each preset dimension as target learning data information; compare the target learning data with the preset standard value to determine the difficulty information of the preset dimension; determine the target difficulty based on the difficulty information, and adjust the test difficulty based on the target difficulty;

[0183] Step S323: Simulate different learning scenarios and learning accuracies, and evaluate students' learning effects at different difficulty levels through simulated scenarios; set an evaluation effect threshold. If the evaluation effect is greater than the threshold, the current test difficulty is too high and needs to be readjusted; if it is less than the threshold, the test difficulty is moderate.

[0184] Preferably, in step S321 of this embodiment, the preset weights corresponding to each preset dimension are obtained, and a comprehensive score is calculated based on the target learning data information and the corresponding preset weights for each preset dimension. By comprehensively considering the learning data and weights of different dimensions, a comprehensive score is obtained that comprehensively reflects the student's learning situation, thus providing a basis for subsequent difficulty adjustments. In step S322, the mean of the initial learning data for each preset dimension is calculated as the target learning data information; the target learning data is compared with the preset standard value to determine the difficulty information for the preset dimension; the target difficulty is determined based on the difficulty information, and the test difficulty is adjusted accordingly. By calculating the mean and comparing the standard value, the difficulty of each dimension is determined, and the test difficulty is adjusted accordingly to ensure that the test accurately reflects the student's learning level. In step S323, different learning scenarios and learning accuracy levels are simulated to evaluate the student's learning performance at different difficulty levels through the simulated scenarios. A threshold for the evaluation effect is set. If the evaluation effect exceeds the threshold, the current test difficulty is considered too high and is adjusted; if it is less than the threshold, the test difficulty is considered moderate. By simulating different scenarios and accuracy levels, the student's learning performance at different difficulty levels is evaluated, and the test difficulty is adjusted based on the evaluation results to ensure that the test achieves the expected results.

[0185] Therefore, this embodiment constitutes a comprehensive, dynamic and efficient evaluation system that can effectively improve the accuracy and adaptability of educational evaluation.

[0186] Further, if Figure 11 As shown, this embodiment also provides a hybrid teaching control system for public courses in colleges and universities. In this embodiment, the hybrid teaching control system for public courses in colleges and universities is applied to the hybrid teaching control method for public courses in colleges and universities in the above-mentioned embodiment. The efficient hybrid teaching system for public courses includes:

[0187] Terminal module 1 is responsible for enabling the user terminal to select the teacher terminal or the student terminal through registration information; granting different permissions based on the teacher terminal or the student terminal; and selecting different modes based on the permissions; after the teacher terminal and the student terminal have completed the mode selection, issuing a student data collection instruction to collect student data; and outputting the collected student data to the prediction module;

[0188] Among them, permissions include teacher side and student side; mode includes teaching method or learning method; teacher side permission allocation selects online teaching or online and offline teaching mode, while student side permission allocation selects online learning or offline learning mode;

[0189] Prediction module 2 is responsible for cleaning and converting the collected student data; statistically summarizing the converted data; performing indicator management and quality management on the summarized student data; predicting student data based on the indicator management and quality management; and outputting the prediction results to the push module;

[0190] Push module 3 is responsible for analyzing student data prediction results, assigning classification labels, and uploading student data to the chain; adjusting the difficulty of the student data of the target students obtained from the blockchain; and pushing courses based on the test results.

[0191] Preferably, the terminal module 1 of this embodiment enables the system to flexibly adapt to the needs of different users and improve the user experience. At the same time, through permission allocation, the security and privacy of the data can be guaranteed to prevent unauthorized access. The data cleaning and conversion of the prediction module 2 are important steps to ensure data quality and accuracy, which directly affect the final analysis results and decisions. Through indicator management and quality management, the accuracy and reliability of the prediction can be improved, thereby providing strong support for educational decision-making. The application of blockchain technology in the push module 3 makes educational records more credible and the educational ecosystem more open. Through on-chain processing and difficulty adjustment testing, the personalization and adaptability of the course content can be guaranteed, and the learning effect and satisfaction of students can be improved.

[0192] Therefore, this embodiment constitutes a complete education system, which realizes the whole process management from user selection to data processing to course push through the technical features of the grid, and improves the intelligence and personalization level of the education system.

[0193] like Figure 12 As shown, this embodiment provides an embodiment of an electronic device. In this embodiment, the electronic device 4 includes a processor 41 and a memory 42 coupled to the processor 41.

[0194] The memory 42 stores program instructions for implementing the hybrid teaching control method for public courses in colleges and universities according to any of the above embodiments.

[0195] The processor 41 is used to execute program instructions stored in the memory 42 to control hybrid teaching of public courses in universities.

[0196] The processor 41 may also be referred to as a CPU (Central Processing Unit). The processor 41 may be an integrated circuit chip having signal processing capabilities. The processor 41 may also be a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic device, a discrete gate or transistor logic device, or a discrete hardware component. The general-purpose processor may be a microprocessor or any conventional processor.

[0197] Furthermore, Figure 13This is a schematic diagram of the structure of a storage medium in an embodiment of the present application. The storage medium 5 in the embodiment of the present application stores program instructions 51 that can implement all the above methods, wherein the program instructions 51 can be stored in the above storage medium in the form of a software product, including a number of instructions for causing a computer device (which can be a personal computer, server, or network device, etc.) or a processor to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, or terminal devices such as a computer, server, mobile phone, and tablet.

[0198] In the several embodiments provided by the present invention, it should be understood that the disclosed systems, devices, and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of units is merely a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interface, device or unit, which can be electrical, mechanical or other forms.

[0199] In addition, the functional units in the various embodiments of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated units may be implemented in the form of hardware or in the form of software functional units. The above is only an embodiment of the present invention and does not limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

[0200] The above detailed description of the specific embodiments of the invention is intended to be illustrative only, and the present invention is not limited to the specific embodiments described above. For those skilled in the art, any equivalent modifications or substitutions to the invention are also within the scope of the present invention. Therefore, equivalent changes, modifications, and improvements made without departing from the spirit and scope of the present invention are also encompassed within the scope of the present invention.

Claims

1. A hybrid teaching control method for public courses in colleges and universities, characterized by: The hybrid teaching control method for public courses in colleges and universities includes: The user terminal selects the teacher terminal or the student terminal through registration information; different permissions are granted according to the teacher terminal or the student terminal; different modes are selected according to the different permissions; after the teacher terminal and the student terminal select the mode, a student data collection instruction is issued to collect student data; Clean and convert the collected student data; perform statistical summarization on the converted data; perform indicator management and quality management on the summarized student data; and make student data predictions based on the indicator management and quality management; Analyze student data prediction results, assign classification labels, and upload student data to the chain; adjust the difficulty of the test on the target student data obtained from the blockchain; and push courses based on the test results.

2. The hybrid teaching control method for public courses in colleges and universities according to claim 1 is characterized in that: The process for issuing instructions for collecting student data includes: The user terminal selects the teacher terminal or the student terminal by filling in information; the device receives the information submitted by the user terminal; and classifies the information into teacher terminal information and student terminal information according to the information classification rules; According to the divided teacher-side information and student-side information, the corresponding account permissions are allocated in a mode; the teacher-side account corresponding to the information is allocated to online teaching or online and offline teaching mode, and the student-side account corresponding to the information is allocated to online learning or offline learning mode; Among them, when the teacher chooses the online teaching mode, the student can only choose the online learning mode; When the teacher and student sides have completed their selections, a collection instruction is issued to collect student learning data and teacher classroom teaching data, and the collected teacher classroom teaching data is stored in the teaching database.

3. The hybrid teaching control method for public courses in colleges and universities according to claim 1 is characterized in that: The process of predicting student data includes: Clean and transform the data collected in the teaching database, analyze and count the processed data to obtain a student data analysis table, and transfer the obtained student data analysis table to the knowledge database; The knowledge database receives the student data analysis table, extracts the student data from the student data analysis table as knowledge information, and stores the knowledge information in the knowledge base; extracts the analyzed and statistically analyzed student data from the student analysis table as analysis information, and stores the analysis information in the analysis database; Among them, knowledge information includes student-related knowledge and document data; analytical data includes homework completion details, extracurricular check-in details, and test score details, representing personal data related to student learning; Preset indicators, read the analysis information corresponding to the indicators from the analysis database, and form indicator statistical data based on the read analysis information; train students to learn prediction models based on the indicator statistical data.

4. The hybrid teaching control method for public courses in colleges and universities according to claim 3 is characterized in that: The process of obtaining the student data analysis table includes: Clean the data in the collected data database, delete outliers, and interpolate missing values; convert unstructured data into structured data and semi-structured data into structured data; The conversion of unstructured data into structured data includes converting source data information in the teaching database into row data or field data and the corresponding relationship between the row data and field data, and writing the row data, field data and the corresponding relationship into the database; Extract key features from the converted data and divide it according to the extracted key feature data; set division criteria and obtain different learning data types according to the division criteria; continuously optimize and adjust according to new data; The classification criteria include students' academic performance, study habits, and participation in extracurricular activities; Statistics are collected on different types of learning data to obtain a student data analysis table, which is then transferred to the knowledge database.

5. The hybrid teaching control method for public courses in colleges and universities according to claim 3 is characterized in that: The process of storing analytical information in the analytical database includes: The student data analysis table extracts personal data related to student learning, including knowledge, document data, homework completion details, extracurricular check-in details, and test score details; Integrate the extracted data, clean the extracted and integrated data, remove invalid data, duplicate data and outliers; detect outliers by comparing data from different sources, and transform the cleaned data; After cleaning and converting the data, knowledge information and analytical information are separated, the knowledge information is stored in the knowledge base, and the analytical information is stored in the analytical database.

6. The hybrid teaching control method for public courses in colleges and universities according to claim 3 is characterized in that: The process of training students to learn predictive models includes: Preset indicators and read the analysis information corresponding to the indicators from the analysis database; generate indicator statistics based on the read analysis information; preset quality inspection standards, calibrate the indicator statistics, and eliminate indicator statistics with quality lower than the preset quality standard to obtain a standardized learning data set; The standardized learning data set is divided into training set, validation set and test set in a ratio of 7:2:1 according to the indicator standard, and the student prediction model is constructed; The student prediction model is trained according to the training set. The training set is input into the first block. The first block outputs the forward prediction result and the backward prediction result. The difference between the input data and the backward prediction result of the previous block is input into the next block, and then the forward prediction result and the backward prediction result of all blocks are obtained. The forward prediction result is integrated to obtain the stacked prediction result. The different stacked prediction results are added together to obtain the global prediction result; the final student learning prediction model is obtained.

7. The hybrid teaching control method for public courses in colleges and universities according to claim 1 is characterized in that: The process of pushing courses based on test results includes: Input student learning data into the prediction model to obtain student data prediction results; construct multiple type labels, extract features from student data prediction results, classify the extracted features according to the type labels, and upload the type labels to the chain; Adjust the difficulty test of student data based on the type label after being uploaded to the chain, simulate different learning scenarios and learning progress, evaluate students' learning effects at different levels of difficulty, set the evaluation effect threshold, and readjust if it is greater than the threshold; if it is less than the threshold, the difficulty is moderate; Analyze the students' prediction results and the difficulty adjustment test results to obtain push keywords; retrieve courses from the knowledge base based on the push keywords, and push the retrieved courses from the knowledge base to students; the teacher receives the push keywords from the student side and answers them online.

8. The hybrid teaching control method for public courses in colleges and universities according to claim 7 is characterized in that: The process of chaining type tags includes: Based on the predicted results of student data, we construct various types of labels and classify the predicted results of student data according to different learning status and learning effects; based on the classification results, we build a student learning management platform based on the blockchain network; Integrate historical student learning data on the student learning management platform; store the student learning data integrated into the student learning management platform on the blockchain to form a trusted data record; compile smart contracts on the student learning management platform and configure the smart contracts to execute on the blockchain network; Manage students’ learning process and progress through smart contracts, and record students’ learning process and progress; Build a student learning management platform based on the blockchain network, construct an intelligent development environment, set up a data interface for the student learning management platform, and integrate existing student learning data through the data interface; The type tags received in real time are configured with uniquely coded identification information. The student learning management platform encrypts the student's learning status and learning results and stores them on the blockchain network. This completes on-chain storage and obtains a credible data record of the student's learning. Obtain students' learning needs, write smart contracts in the intelligent development environment of the student learning management platform, and deploy the written smart contracts to the blockchain network; Among them, students’ learning needs include improving learning efficiency, improving scores, and solving specific learning problems.

9. The hybrid teaching control method for public courses in colleges and universities according to claim 7 is characterized in that: The process of evaluating students' learning outcomes at different levels of difficulty includes: Obtain the preset weight corresponding to each preset dimension, and calculate a comprehensive score based on the target learning data information on each preset dimension and the corresponding preset weight; Among them, the preset dimensions include learning time, answer accuracy and learning progress; the target learning data information includes historical scores and current learning progress; Calculate the mean of the initial learning data on each preset dimension as the target learning data information; compare the target learning data with the preset standard value to determine the difficulty information of the preset dimension; determine the target difficulty based on the difficulty information, and adjust the test difficulty based on the target difficulty; Simulate different learning scenarios and learning accuracy, and evaluate students' learning effects at different difficulty levels through simulated scenarios; set an evaluation effect threshold. If the evaluation effect is greater than the threshold, the current test difficulty is too high and needs to be readjusted; if it is less than the threshold, the test difficulty is moderate.

10. A hybrid teaching control system for public courses in colleges and universities, which is applied to the efficient hybrid teaching method for public courses as claimed in any one of claims 1 to 9, characterized in that: include: The terminal module is responsible for enabling the user terminal to select the teacher terminal or the student terminal through registration information; granting different permissions based on the teacher terminal or the student terminal; and selecting different modes based on the permissions. After the teacher terminal and the student terminal have completed the mode selection, they issue a student data collection instruction to collect student data; and output the collected student data to the prediction module. Among them, permissions include teacher side and student side; mode includes teaching method or learning method; teacher side permission allocation selects online teaching or online and offline teaching mode, while student side permission allocation selects online learning or offline learning mode; The prediction module is responsible for cleaning and converting the collected student data; performing statistical summarization on the converted data; performing indicator management and quality management on the summarized student data; predicting student data based on the indicator management and quality management; and outputting the prediction results to the push module; The push module is responsible for analyzing student data prediction results, assigning classification labels, and uploading student data to the chain; adjusting the difficulty of the student data of the target students obtained from the blockchain; and pushing courses based on the test results.

Images