Online teaching method based on big data
By collecting and analyzing students' multidimensional learning data, adjusting teaching strategies and prediction models in real time, and providing personalized intervention and feedback, the problem of insufficient real-time response, personalization and prediction capabilities of the existing online teaching system is solved, and the teaching effect is improved and Student experience.
Patent Information
- Application Number
- CN202411465959.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-10-21
- Publication Date
- 2024-12-06
- Estimated Expiration
- 2044-10-21
AI Technical Summary
The existing online teaching system lacks real-time responsiveness, insufficient personalization, and limited prediction capabilities, which increases the burden on teachers and cannot effectively support students' personalized learning needs and real-time feedback.
By collecting students' multidimensional learning data, processing and analysis in real time, generating learning status assessments, adjusting teaching strategies, building learning behavior prediction models, providing personalized interventions and real-time feedback, and optimizing model parameters using supervised learning algorithms.
Real-time assessment and dynamic adjustment of students' learning status are achieved, personalized teaching is provided, early warning and preventive intervention are provided, and teaching effect and student experience are improved.
Smart Images

Figure CN119091704A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of online teaching, and in particular to an online teaching method based on big data. Background Art
[0002] With the rapid development of information technology, online education has gradually become an important part of modern education. However, traditional online teaching methods often lack personalized learning support for individual students. Traditional online teaching mainly relies on standardized course content and a fixed teaching rhythm, and cannot respond to students' learning needs and feedback in real time. This method may cause some students to fall behind due to inability to keep up with the course progress, and may also lead to a waste of learning resources, because not all students need the same learning content and teaching methods. In addition, it is difficult for teachers to accurately grasp the learning status and understanding of each student, and they cannot provide effective help and guidance in a timely manner.
[0003] The existing online teaching systems have the following main problems: lack of real-time response capabilities and inability to quickly adjust teaching strategies; insufficient personalization, only simple content adjustments, and inability to deeply understand students' learning behaviors; limited predictive capabilities, making it difficult to accurately identify and provide early warning of difficulties that students may encounter; in addition, these systems increase the burden on teachers, requiring manual analysis of students' learning situations and adjustment of teaching strategies, which affects teaching efficiency and effectiveness. Summary of the invention
[0004] In view of the shortcomings of the prior art, the present invention provides an online teaching method based on big data, which solves the problems of poor real-time response, insufficient personalization, limited prediction ability and increased burden on teachers in the existing online teaching system.
[0005] To achieve the above objectives, the present invention is implemented through the following technical solutions: an online teaching method based on big data, comprising the following steps:
[0006] S1: Collect students’ learning data D on the online learning platform;
[0007] S2: Process and analyze the learning data D in real time to generate a student's learning status evaluation S;
[0008] S3: Adjust teaching strategy T based on learning status assessment S;
[0009] S4: Build a learning behavior prediction model to predict students’ future learning performance
[0010] S5: Based on the prediction results Generate a warning signal W;
[0011] S6: Provide personalized interventions, including customized learning paths and real-time feedback;
[0012] S7: Use supervised learning algorithms to train learning behavior prediction models and optimize model parameters.
[0013] Preferably, the S1 comprises:
[0014] Collect students’ learning data D = {d1, d2, ..., d n}, including learning progress, engagement, feedback, assignment submissions, quiz scores, course viewing time, and class participation.
[0015] Preferably, S2 includes:
[0016] The learning data D is processed in real time to generate the student's learning status evaluation S, which is expressed by the following formula:
[0017]
[0018] Among them, S i is the learning status evaluation of the ith student, wik is the weight of data type k, dik is the kth data indicator of the ith student, and m is the number of data types;
[0019] The learning data D is standardized, and the standardization formula is:
[0020]
[0021] Among them, d' ik is the standardized data, μk is the mean of data type k, and σk is the standard deviation of data type k.
[0022] Preferably, S3 includes:
[0023] Adjust the teaching strategy T based on the learning status evaluation S. The adjustment formula is:
[0024] T=f(S)
[0025] Where f is the teaching strategy adjustment function, which dynamically adjusts the teaching content and rhythm according to S;
[0026] Dynamically adjust the course content difficulty C based on learning status evaluation S d , the adjustment formula is:
[0027] C d =α1S i +α2
[0028] Among them, α1 and α2 are adjustment coefficients;
[0029] Adjust interaction frequency I f , through the formula:
[0030] I f =γ1S i +γ2
[0031] Where γ1 and γ2 are the interaction frequency adjustment coefficients.
[0032] Preferably, S4 includes:
[0033] Construct a learning behavior prediction model P to predict students' future learning performance The prediction model formula is:
[0034]
[0035] in is the predicted learning performance, β0 is the model constant term, β j For learning status assessment, S j The coefficient of , P is the number of evaluation variables;
[0036] When the predicted value When it is lower than a predetermined threshold value θ, an early warning signal W is generated, where W-1 indicates that the early warning is triggered, and W=0 indicates that the early warning is not triggered.
[0037] Preferably, S5 includes:
[0038] Provide customized learning path L, based on learning status assessment S and predicted results The path selection formula is:
[0039]
[0040] Where ι is the set of all optional learning paths, and u is the path utility function;
[0041] Provide real-time feedback Where h is the feedback generation function, which generates feedback information based on the learning state and prediction results.
[0042] Preferably, the S7 includes:
[0043] Use supervised learning algorithm to train historical learning data and optimize model parameters by minimizing the loss function. The loss function is:
[0044]
[0045] Among them, N is the number of samples, yk is the actual learning result, Predict the results for the model.
[0046] Beneficial Effects
[0047] The present invention provides an online teaching method based on big data. Compared with the prior art, it has the following beneficial effects:
[0048] In the present invention, by collecting and analyzing students' learning data on the online learning platform, real-time evaluation and dynamic adjustment of students' learning status are achieved. The method can timely identify students' problems and difficulties in the learning process, and then provide personalized intervention measures to effectively improve teaching effects and students' learning experience. First, through steps S1 and S2, students' multi-dimensional learning data are collected and analyzed, covering various factors such as learning progress, participation, feedback, homework submission, test scores, course viewing time, and class participation. Using these data, students' learning behavior and status can be fully understood, laying the foundation for personalized teaching. Secondly, in step S3, the teaching strategy is adjusted based on the results of the learning status evaluation, so that the teaching content and rhythm are more in line with the actual needs of students. This dynamic adjustment The teaching method can avoid the one-size-fits-all teaching model and ensure that each student can learn at the pace and difficulty that best suits him or her, which helps to improve learning outcomes. Thirdly, the learning behavior prediction model constructed through steps S4 and S5 can foresee students' future learning performance and generate early warning signals. This function enables teachers to take intervention measures in advance to prevent students from falling behind in learning. Through real-time prediction and early warning, preventive intervention can be carried out for potential learning difficulties to improve the foresight and effectiveness of teaching. Finally, the personalized learning path and real-time feedback provided by steps S6 and S7, combined with the learning behavior prediction model optimized by the supervised learning algorithm, enable the system to provide customized learning plans and instant feedback based on students' personalized needs, further improving students' learning experience and satisfaction. BRIEF DESCRIPTION OF THE DRAWINGS
[0049] Figure 1 This is a flow chart of an online teaching method based on big data proposed by the present invention. DETAILED DESCRIPTION
[0050] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0051] See also Figure 1 The present invention provides two technical solutions, which specifically include the following embodiments:
[0052] Example:
[0053] An online teaching method based on big data includes the following steps:
[0054] S1: Collect students’ learning data D on the online learning platform. The online learning platform automatically collects students’ learning data. These data reflect students’ learning behaviors from multiple dimensions and provide a basis for subsequent analysis and processing.
[0055] S2: Real-time processing and analysis of learning data D to generate students' learning status evaluation S, using data analysis algorithms to process the collected learning data in real time, and generate comprehensive learning status evaluation through weighted summation and other methods. Data standardization eliminates scale differences between different data types;
[0056] S3: Adjust the teaching strategy T based on the learning status evaluation S. Based on the learning status evaluation results, use the adjustment function to dynamically adjust the teaching content and rhythm. Difficulty adjustment and interaction frequency adjustment ensure that the teaching strategy can adapt to the students' actual learning situation;
[0057] S4: Construct a learning behavior prediction model to predict students’ future learning performance Build a learning behavior prediction model to predict students' future learning performance through linear regression or other machine learning algorithms. If the prediction result is lower than the predetermined threshold, the system generates an early warning signal to alert the teacher;
[0058] S5: Based on the prediction results Generates early warning signal W, and the system provides personalized learning paths and real-time feedback based on the prediction results and learning status assessment. The path selection formula and feedback generation formula ensure that each student receives support that meets their learning needs;
[0059] S6: Provide personalized interventions, including customized learning paths and real-time feedback;
[0060] S7: Use supervised learning algorithms to train learning behavior prediction models and optimize model parameters.
[0061] S1 includes:
[0062] Collect students’ learning data D = {d1, d2, ..., d n}, including learning progress, participation, feedback, homework submission, test scores, course viewing time and class participation, among which learning progress, participation, feedback, homework submission, test scores, course viewing time and class participation are as follows:
[0063] Learning progress: course chapters or modules completed by students;
[0064] Participation: students’ participation in discussion forums and interactive sessions;
[0065] Feedback: students’ evaluation of course content or teaching methods;
[0066] Assignment submission status: submission time, completion quality, etc.;
[0067] Test results: scores of regular tests and their changing trends;
[0068] Course viewing time: the time and duration of video viewing;
[0069] Class participation: the activity of the online class, including the number of questions asked and the frequency of interaction;
[0070] S2 includes:
[0071] The learning data D is processed in real time to generate the student's learning status evaluation S, which is expressed by the following formula:
[0072]
[0073] Among them, S i is the learning status evaluation of the ith student, wik is the weight of data type k, dik is the kth data indicator of the ith student, and m is the number of data types;
[0074] The learning data D is standardized, and the standardization formula is:
[0075]
[0076] Among them, d' ik is the standardized data, μk is the mean of data type k, and σk is the standard deviation of data type k.
[0077] S3 includes:
[0078] Adjust the teaching strategy T based on the learning status evaluation S. The adjustment formula is:
[0079] T=f(S)
[0080] Where f is the teaching strategy adjustment function, which dynamically adjusts the teaching content and rhythm according to S;
[0081] Dynamically adjust the course content difficulty C based on learning status evaluation S d , the adjustment formula is:
[0082] C d =α1S i +α2
[0083] Among them, α1 and α2 are adjustment coefficients;
[0084] Adjust interaction frequency I f , through the formula:
[0085] If =γ1S i +γ2
[0086] Where γ1 and γ2 are the interaction frequency adjustment coefficients.
[0087] S4 includes:
[0088] Construct a learning behavior prediction model P to predict students' future learning performance The prediction model formula is:
[0089]
[0090] in is the predicted learning performance, β0 is the model constant term, β j For learning status assessment, S j The coefficient of , P is the number of evaluation variables;
[0091] When the predicted value When it is lower than a predetermined threshold value θ, an early warning signal W is generated, where W-1 indicates that the early warning is triggered, and W=0 indicates that the early warning is not triggered.
[0092] S5 includes:
[0093] Provide customized learning path L, based on learning status assessment S and predicted results The path selection formula is:
[0094]
[0095] Where ι is the set of all optional learning paths, and u is the path utility function;
[0096] Provide real-time feedback Where h is the feedback generation function, which generates feedback information based on the learning state and prediction results.
[0097] S7 includes:
[0098] Use supervised learning algorithm to train historical learning data and optimize model parameters by minimizing the loss function. The loss function is:
[0099]
[0100] Among them, N is the number of samples, yk is the actual learning result, Predict the results for the model.
[0101] When working, it mainly evaluates students' learning status in real time by collecting and analyzing students' behavioral data on the online learning platform, and adjusts teaching strategies accordingly to provide personalized learning experience. The working principle of this method mainly includes five core steps: data collection, data processing, teaching strategy adjustment, learning behavior prediction and personalized intervention. First, in the data collection stage (S1), the system automatically collects students' learning data through the online learning platform. These data include but are not limited to learning progress, participation, feedback, homework submission, test scores, course viewing time and class participation. These data reflect students' learning behaviors from multiple dimensions and provide a basis for subsequent analysis and processing. Next, in the data processing and learning status evaluation stage (S2), the system uses data analysis algorithms to process the collected learning data in real time. Different types of data are combined through methods such as weighted summation to generate a comprehensive learning status evaluation S. The formula is Among them, S i is the learning status evaluation of the i-th student, wik is the weight of data type k, dik is the score of the i-th student, and m is the number of data types. Data standardization is performed through the formula In order to eliminate the scale differences between different data types, in the teaching strategy adjustment stage (S3), the system uses the adjustment function T = f (S) to dynamically adjust the teaching content and rhythm based on the learning state evaluation result S. For difficulty adjustment, the system uses formula C d =α1S i +α2, calculate the difficulty of the course content, and adjust the interaction frequency through formula I f =γ1S i +γ2, these adjustments ensure that the teaching strategy can adapt to the students' actual learning situation and provide personalized teaching content. In the learning behavior prediction stage (S4), the system constructs a learning behavior prediction model P to predict students' future learning performance. By using linear regression or other machine learning algorithms, the model calculates the predicted value in is the predicted learning performance, β0 is the model constant term, β j For learning status assessment, S j The coefficient of P is the number of evaluation variables. If the prediction result When it is lower than the predetermined threshold θ, the system will generate an early warning signal W to remind the teacher. Finally, in the personalized intervention stage (S5 and S6), the system provides personalized learning paths and real-time feedback based on the prediction results and learning status evaluation. The path selection formula is: The real-time feedback generation formula is: Ensure that every student can get support that meets their learning needs. Optimize model parameters through supervised learning algorithms to further improve the accuracy and reliability of the prediction model, thereby continuously improving personalized teaching effects. In short, the present invention integrates big data analysis, machine learning and educational technology to build an intelligent online teaching system, realize personalized teaching management and real-time learning status monitoring, and provide comprehensive technical support for online education.
[0102] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the application should be included in the protection scope of the present application.
Claims
1. An online teaching method based on big data, Features: The following steps are involved: S1: Collect students’ learning data D on the online learning platform; S2: Process and analyze the learning data D in real time to generate a student's learning status evaluation S; S3: Adjust teaching strategy T based on learning status assessment S; S4: Build a learning behavior prediction model to predict students’ future learning performance S5: Based on the prediction results Generate a warning signal W; S6: Provide personalized interventions, including customized learning paths and real-time feedback; S7: Use supervised learning algorithms to train learning behavior prediction models and optimize model parameters.
2. According to the online teaching method based on big data in claim 1, Features: The S1 includes: Collect students’ learning data D = {d 1 ,d 2 ,......,d n }, including learning progress, engagement, feedback, assignment submissions, quiz scores, course viewing time, and class participation.
3. According to the online teaching method based on big data in claim 1, Features: The S2 includes: The learning data D is processed in real time to generate the student's learning status evaluation S, which is expressed by the following formula: Among them, S i is the learning status evaluation of the ith student, wik is the weight of data type k, dik is the kth data indicator of the ith student, and m is the number of data types; The learning data D is standardized, and the standardization formula is: Among them, d' ik is the standardized data, μk is the mean of data type k, and σk is the standard deviation of data type k.
4. According to the online teaching method based on big data as claimed in claim 1, Features: The S3 includes: Adjust the teaching strategy T based on the learning status evaluation S. The adjustment formula is: T=f(S) Where f is the teaching strategy adjustment function, which dynamically adjusts the teaching content and rhythm according to S; Dynamically adjust the course content difficulty C based on learning status evaluation S d , the adjustment formula is: C d =a 1 S i +a 2 where α 1 and α 2 is the adjustment factor; Adjust interaction frequency I f , through the formula: I f =c 1 S i +g 2 where γ 1 and γ 2 is the interaction frequency adjustment factor.
5. According to the online teaching method based on big data as claimed in claim 1, Features: The S4 includes: Construct a learning behavior prediction model P to predict students' future learning performance The prediction model formula is: in is the predicted learning performance, β 0 is the model constant term, β j For learning status assessment, S j The coefficient of , P is the number of evaluation variables; When the predicted value When it is lower than a predetermined threshold value θ, an early warning signal W is generated, where W-1 indicates that the early warning is triggered, and W=0 indicates that the early warning is not triggered.
6. According to the online teaching method based on big data in claim 1, Features: The S5 includes: Provide customized learning path L, based on learning status assessment S and predicted results The path selection formula is: Where ι is the set of all optional learning paths, and u is the path utility function; Provide real-time feedback Where h is the feedback generation function, which generates feedback information based on the learning state and prediction results.
7. According to the online teaching method based on big data in claim 1, Features: The S7 includes: Use supervised learning algorithm to train historical learning data and optimize model parameters by minimizing the loss function. The loss function is: Among them, N is the number of samples, yk is the actual learning result, Predict the results for the model.
Citation Information
Patent Citations
An educational big data analysis method based on artificial intelligence
AU2020103529A4
Learning early warning method and device based on multi-feature modeling and multi-level evaluation
CN116308935A
Personalized education method based on artificial intelligence
CN117852758A
Automatic interactive intelligent education method and system
CN118537175A
Teaching quality evaluation system based on big data analysis
CN118569723A
Cited By
English learning system and method based on mobile device
CN119474687A
A smart digital human control system and method
CN119762307A
Education effect evaluation method and system based on artificial intelligence
CN119831186A
Intelligent teaching method, system, device and equipment and readable storage medium
CN120182061A
