Teaching content dynamic adjustment system and method based on education assistance

Abstract

The invention discloses a teaching content dynamic adjustment system and method based on education assistance, and relates to the field of education assistance, and the system comprises a global perception module, an intelligent analysis module, a decision generation module, and a teaching content updating module. The global sensing module is used for capturing behavior details and environment data of students in a classroom and learning tracks of the students on a learning platform, and the intelligent analysis module is used for capturing behaviors of the students in the classroom and environment data of the classroom, analyzing a current teaching effect based on the captured data and sending the analyzed current teaching effect to the classroom. And an external learning platform is fused for mastering student learning. According to the teaching content dynamic adjustment system and method based on education assistance, teaching adjustment strategy suggestions can be provided for teachers, potential relations between knowledge points of different subjects can be mined through the knowledge association graph, interdisciplinary fusion teaching strategies can be made, and the comprehensive thinking ability of students can be cultivated.

Description

Technical Field

[0001] The present invention relates to the field of educational assistance, and in particular to a system and method for dynamically adjusting teaching content based on educational assistance. Background Art

[0002] Teaching is a unique talent cultivation activity consisting of teachers’ teaching and students’ learning. Through this activity, teachers guide students to learn and master cultural and scientific knowledge and skills in a purposeful, planned and organized manner, promote the improvement of students’ quality and make them become people needed by society.

[0003] When providing educational assistance, due to the limitations of data collection and analysis, it is impossible to accurately grasp the learning needs of each student, resulting in a single teaching method that is difficult to meet the learning progress and ability differences of different students. Traditional teaching content and methods are difficult to arouse students' interest and emotional attention, which can easily cause students to have learning fatigue and lose their motivation to learn. It is also difficult to integrate subjects, which is not conducive to cultivating students' comprehensive literacy and ability to solve practical problems.

[0004] Therefore, it is necessary to propose a dynamic adjustment system and method of teaching content based on educational assistance to solve the above problems. Summary of the invention

[0005] The main purpose of the present invention is to provide a system and method for dynamically adjusting teaching content based on educational assistance, which can effectively solve the problems in the background technology.

[0006] To achieve the above object, the technical solution adopted by the present invention is:

[0007] A teaching content dynamic adjustment system based on educational assistance, comprising a global perception module, an intelligent analysis module, a decision generation module, and a teaching content update module. The global perception module is used to capture students' behavioral details, environmental data, and learning trajectories on a learning platform in a classroom, and to judge teaching content based on the captured data, while grasping students' learning behavior.

[0008] The intelligent analysis module is used to capture the behavior of students in the classroom and the environmental data of the classroom, analyze the current teaching effect based on the captured data, and integrate the external learning platform to grasp the students' learning;

[0009] The decision generation module builds a unique personalized learning model for each student, and provides teachers with interactive feedback on teaching adjustment strategy suggestions based on the personalized learning model;

[0010] The teaching content updating module rearranges and optimizes the teacher's teaching content based on the decision generated by the decision generating module.

[0011] Preferably, the global perception module includes a behavior detail capture module, an environmental data integration module, and a learning platform deep fusion module, wherein the behavior detail capture module is based on the arrangement of high-definition cameras and precise sound pickup equipment in the classroom, and is used to record students' obvious behaviors in the classroom, including but not limited to speaking, raising hands to answer, and also to capture students' subtle behaviors in the classroom, including but not limited to the direction of eye focus, pauses and hesitations when writing. When the intelligent analysis module detects that most students have wandering eyes when explaining a certain knowledge point, it means that the students' attention is distracted, and it is judged that the effect of explaining the knowledge point is not good at this time;

[0012] The environmental data integration module collects classroom environmental data in all directions based on sensors, including temperature, humidity, light intensity, noise decibels, and analyzes it in combination with teaching time and course type;

[0013] The learning platform deep integration module is used to deeply integrate with various online learning platforms, electronic textbooks, and homework systems to collect students' learning trajectories on different platforms, including homework completion status, test scores, annotation content on electronic textbooks, and download preferences for learning materials, so as to fully understand students' learning behavior.

[0014] Preferably, the intelligent analysis module includes a cognitive and emotional dual-dimensional analysis module, a knowledge association map dynamic update module, and a personalized learning model precision construction module, wherein the cognitive and emotional dual-dimensional analysis module includes a multimodal data fusion analysis module, a knowledge transfer and cognitive network construction module, an emotional fluctuation prediction and context association module, a value and interest tendency mining module, and an emotional resonance and cognitive guidance strategy generation module. The multimodal data fusion analysis module is used to fuse and analyze text data, voice, expression, and body movements. When students answer questions, their voice intonation, speaking speed, facial expressions, and body movements are combined to more comprehensively judge and analyze their mastery of knowledge and their current emotional state;

[0015] The knowledge transfer and cognitive network construction module is used to deeply analyze the students' knowledge transfer between different subjects and different knowledge points, and based on this, build a unique cognitive network for each student to demonstrate their mastery of each knowledge point and the strength of the association between knowledge points and the students' thinking path when applying knowledge. It is used to discover the weak links and potential advantages in the students' cognitive structure and provide an accurate basis for personalized learning.

[0016] Preferably, the emotion fluctuation prediction and context association module dynamically tracks and predicts the students' emotional state based on time series analysis and context awareness technology, and adjusts the teaching method according to the tracking and prediction structure;

[0017] The value and interest mining module mines students' potential values ​​and interest tendencies based on their opinions expressed in class discussions, homework and projects, and their topic preferences. It combines these mining results with teaching content to make teaching more targeted and attractive, and is used to stimulate students' intrinsic learning motivation.

[0018] The emotional resonance and cognitive guidance strategy generation module generates personalized emotional resonance and cognitive guidance strategies based on a dual-dimensional analysis of students' cognition and emotions, and is used to provide teachers with strategic suggestions based on the characteristics of different students to achieve the best teaching results.

[0019] Preferably, the knowledge association map dynamic update module is used to construct and update the knowledge association map of each subject in real time. The knowledge association map is used to display the logical relationship between knowledge points and mark each student's mastery of different knowledge points and learning path based on the students' learning data.

[0020] Preferably, the personalized learning model precision construction module includes a multi-source data integration module, a feature extraction and analysis module, and a model construction and verification module. The multi-source data integration module is used to collect students' learning behavior data, academic performance data, classroom interaction data, and self-assessment data, wherein the learning behavior data includes students' operation records on various learning platforms; the academic performance data includes daily homework, periodic tests, and test scores, as well as the scores of each knowledge point and the distribution of answering time; the classroom interaction data is used to record students' speech content, question frequency, activeness in participating in group discussions, and interaction methods with classmates and teachers in class; the self-assessment data includes regularly collecting students' self-assessment questionnaires on their own learning status, interests, and goals;

[0021] The feature extraction and analysis module is used to extract and analyze students' cognitive ability, learning style, emotional attitude, and interest preference. Cognitive ability is based on academic performance and learning behavior data to analyze students' memory ability, logical reasoning ability, and spatial imagination ability; learning style is determined according to students' preferences for learning materials and classroom interaction methods; emotional attitude is used to combine classroom interaction and self-assessment data to judge students' learning motivation, self-confidence, anxiety and other emotional factors; interest preference is based on students' self-selected extended learning content, participating community activities, and topics of concern in discussions. Students' cognitive ability is scored by the formula:

[0022]

[0023] Sc is the student’s comprehensive cognitive ability score; is the student's score on the i-th cognitive ability dimension; n is the number of cognitive ability dimensions; w iis the weight of the i-th cognitive ability dimension. Cognitive ability is divided into excellent: 85-100 points; good: 70-84 points; medium: 60-69 points; poor: below 60 points;

[0024] The model building and verification module is based on a neural network model, and uses the extracted features as input to build a learning model unique to each student. The learning model is used to predict the student's performance in future learning tasks. As new data is continuously input, the learning model adjusts parameters in real time to achieve dynamic updates. The learning model is regularly verified through actual teaching results, and the learning model prediction results are compared with the students' actual learning performance. If the prediction deviation is large, the cause is analyzed and the model parameters and feature selection methods are adjusted to ensure the accuracy and reliability of the model.

[0025] Preferably, the decision generation module includes a human-computer collaborative teaching decision module, an interdisciplinary integration strategy formulation module, and a teaching plan optimization module. The human-computer collaborative teaching decision module includes a suggestion diversification module, a teacher-led optimization module, and a real-time communication feedback module. The suggestion diversification module provides teachers with a variety of teaching adjustment strategy suggestions based on the personalized learning model constructed for students.

[0026] The teacher-led optimization module comprehensively evaluates and optimizes the suggestions provided by the suggestion diversification module based on the teacher's rich teaching experience, in-depth understanding of individual students and teaching goals. The teacher can adjust the suggestions based on the overall atmosphere of the class and the interpersonal relationships between students;

[0027] The real-time communication feedback module is used to maintain real-time communication between the teacher and the human-computer collaborative teaching decision-making module during the teaching decision-making process. The teacher can raise questions about the suggestions provided by the human-computer collaborative teaching decision-making module, and the human-computer collaborative teaching decision-making module will immediately provide explanations and additional explanations to promote further data analysis and provide more accurate suggestions for subsequent teaching decisions.

[0028] Preferably, the interdisciplinary integration strategy formulation module analyzes the explicit connections between knowledge points of different disciplines based on the constructed knowledge association map and explores the potential deep-level connections, and formulates interdisciplinary integration teaching strategies based on this, so as to enable students to understand the occurrence and development of knowledge from multiple perspectives; takes complex problems and hot topics in real life as the theme, designs interdisciplinary integration teaching projects, so as to enable students to learn to use multidisciplinary knowledge to solve problems; and at the same time builds an interdisciplinary teacher communication and collaboration platform, on which teachers of different disciplines can share teaching experiences, discuss integrated teaching strategies, and jointly design teaching content and evaluation methods for students;

[0029] The teaching plan optimization module is used to track students' learning status in real time, and dynamically adjust the teaching plan according to students' immediate feedback in class, homework completion status and periodic test results; based on students' personalized learning model and learning progress, a long-term learning path plan is formulated for students, covering the learning sequence and depth of knowledge in various subjects, as well as students' interest development and future career direction; for emergencies in the teaching process, including students' poor learning status due to special events and the school's temporary adjustment of teaching arrangements, the impact of the event on students' learning is quickly analyzed, and corresponding response decisions are provided. Based on the autoregressive integrated sliding average model, the learning progress of students is predicted. Assume that the time series y t represents the student’s learning index at time t, and the model expression is:

[0030] Θ(B)(1-B) d y t =Φ(B)∈ t ;

[0031] Where B is the backward shift operator; Θ(B) is the autoregressive part; Φ(B) is the sliding average part; ∈ t is a white noise sequence; d is the difference order; based on this, the autoregressive order p and the sliding average order q are determined, and the learning progress of future students is predicted by fitting the historical learning data.

[0032] Preferably, the teaching content updating module automatically rearranges and optimizes the teaching content based on the teaching decision, selects appropriate multimedia materials such as pictures, videos, and audios from the teaching resource library, organically integrates them with the original teaching content, and generates new teaching courseware.

[0033] A method for dynamically adjusting teaching content based on educational assistance includes the following steps:

[0034] S1: Multi-source data collection and analysis: collect students’ learning behavior data, academic performance data, classroom interaction data, and self-assessment data in the classroom, and use machine learning algorithms to deeply analyze these data to extract students’ cognitive ability characteristics, learning style characteristics, emotional attitude characteristics, and interest preference characteristics;

[0035] S2: Personalized learning model construction. Analyze the collected features and use the neural network model to build a unique learning model for each student to predict the student's future learning performance. Update and optimize in real time with new data input. At the same time, regularly verify the model through actual teaching results and calibrate deviations.

[0036] S3: Human-computer collaborative teaching decision-making, based on students' characteristic data and personalized learning models, provides teachers with a variety of teaching adjustment strategy suggestions. The teaching adjustment strategy suggestions include theoretical basis, expected effects and potential challenges. Teachers evaluate and optimize the suggestions based on their own teaching experience, understanding of individual students and teaching goals, and adjust the strategies according to the actual situation of the class. During the teaching decision-making process, teachers can ask questions about the suggestions; teachers can also provide feedback on problems found in practice, which is used to prompt them to further analyze and provide more accurate suggestions;

[0037] S4: Formulate interdisciplinary integration strategies, use knowledge association maps to deeply explore the explicit and potential connections between knowledge points in different disciplines, design interdisciplinary integration teaching projects based on real-life problems and hot topics, build an interdisciplinary teacher communication and collaboration platform, promote teachers from different disciplines to jointly design teaching content and evaluation methods, and achieve complementary disciplinary advantages;

[0038] S5: Dynamic teaching plan optimization, real-time tracking of students' classroom expressions, questions, participation, homework, and test results, based on which the teaching rhythm and content are dynamically adjusted. At the same time, based on students' personalized learning models and progress, long-term learning path plans are formulated, taking into account interests and career directions. In the face of teaching emergencies, the impact can be quickly analyzed and response decisions can be provided to ensure the achievement of teaching goals.

[0039] Compared with the prior art, the present invention provides a system and method for dynamically adjusting teaching content based on educational assistance, which has the following beneficial effects:

[0040] 1. The system and method for dynamic adjustment of teaching content based on educational assistance can not only collect traditional academic performance and homework completion data, but also integrate students' facial expressions, body language, voice intonation, and classroom environment data in class. Based on this and combined with the current teaching content, it can judge the difficulties students have in understanding knowledge. At the same time, through knowledge graph technology, it can build a student's knowledge mastery network, clearly present the relationship between knowledge points and the students' mastery of each knowledge point.

[0041] 2. This dynamic adjustment system and method of teaching content based on educational assistance can tailor a personalized learning model for each student based on the extracted multi-dimensional characteristics of students, including cognitive ability, learning style, emotional attitude, and interest preferences. When the student's learning methods and learning effects change at a certain stage, the model can reflect these changes in a timely manner and provide a basis for teaching decisions that is more in line with the student's current status.

[0042] 3. The dynamic adjustment system and method of teaching content based on educational assistance can provide teachers with a rich variety of teaching adjustment strategy suggestions. Teachers can evaluate, optimize and make final decisions on the suggestions based on their own teaching experience, understanding of students and teaching objectives. By using knowledge association maps to explore the potential connections between knowledge points in different disciplines, they can formulate interdisciplinary integrated teaching strategies and cultivate students' comprehensive thinking ability. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 It is a system block diagram of the present invention. DETAILED DESCRIPTION

[0044] In order to make the technical means, creative features, objectives and effects achieved by the present invention easy to understand, the present invention is further explained below in conjunction with specific implementation methods.

[0045] Embodiment 1:

[0046] like Figure 1 As shown, a teaching content dynamic adjustment system based on educational assistance includes a global perception module, an intelligent analysis module, a decision-making module, and a teaching content update module. The global perception module is used to capture students' behavioral details, environmental data, and learning trajectories on the learning platform in the classroom, and to judge the teaching content based on the captured data, while grasping the students' learning behavior.

[0047] The global perception module includes a behavior detail capture module, an environmental data integration module, and a learning platform deep integration module. The behavior detail capture module is based on high-definition cameras and precise sound pickup equipment arranged in the classroom to record students' obvious behaviors in class, including but not limited to speaking and raising hands to answer questions. It is also used to capture students' subtle behaviors in class, including but not limited to the direction of eye focus and pauses and hesitations when writing. When the intelligent analysis module detects that most students' eyes are wandering when explaining a certain knowledge point, it means that the students' attention is distracted, and it is judged that the effect of explaining the knowledge point is not good at this time;

[0048] The environmental data integration module collects classroom environmental data in all aspects based on sensors, including temperature, humidity, light intensity, and noise decibels, and analyzes it in combination with teaching time and course type. When it is found that the temperature in the classroom rises in the afternoon, the accuracy and enthusiasm of students in answering questions decrease, the teacher can be reminded to increase the break time between classes and adjust the teaching rhythm in the summer afternoon.

[0049] The learning platform deep integration module is used to deeply integrate with various online learning platforms, electronic textbooks, and homework systems to collect students' learning trajectories on different platforms, including homework completion status, test scores, annotations on electronic textbooks, and download preferences for learning materials, in order to fully understand students' learning behaviors.

[0050] The intelligent analysis module is used to capture students' behaviors and classroom environment data in the classroom, analyze the current teaching effect based on the captured data, and integrate the external learning platform to grasp students' learning;

[0051] The intelligent analysis module includes a cognitive and emotional dual-dimensional analysis module, a knowledge association map dynamic update module, and a personalized learning model precision construction module. The cognitive and emotional dual-dimensional analysis module includes a multimodal data fusion analysis module, a knowledge transfer and cognitive network construction module, an emotional fluctuation prediction and context association module, a value and interest tendency mining module, and an emotional resonance and cognitive guidance strategy generation module. The multimodal data fusion analysis module is used to integrate and analyze text data, voice, expression, and body movements. When students answer questions, their voice intonation, speaking speed, facial expressions, and body movements are combined to more comprehensively judge and analyze their knowledge mastery and current emotional state.

[0052] The knowledge transfer and cognitive network construction module is used to deeply analyze students' knowledge transfer between different subjects and different knowledge points. For example, it observes whether the logical reasoning methods mastered by students in mathematics can be transferred to the solution of physical problems. Based on this, a unique cognitive network is constructed for each student to demonstrate their mastery of each knowledge point and the strength of the association between knowledge points and the students' thinking path when applying knowledge. It is used to discover the weak links and potential advantages in students' cognitive structure, providing accurate basis for personalized learning.

[0053] The emotional fluctuation prediction and situational association module dynamically tracks and predicts students' emotional states based on time series analysis and situational awareness technology, and adjusts teaching methods based on the tracking and prediction structure. For example, it combines the contextual factors of course content, teaching activities, and students' learning experience of the day to predict students' possible emotional fluctuations in the following learning process. If a student does not perform well in a test of a course in the morning, and is studying another course in the afternoon, it is predicted that he or she may lack confidence and become anxious, and the teacher is reminded in advance to pay attention and adjust the teaching method.

[0054] The module for mining values ​​and interests explores students' potential values ​​and interests based on their opinions and topic preferences in classroom discussions, homework and projects. It combines these mining results with teaching content to make teaching more targeted and attractive, and to stimulate students' intrinsic learning motivation. For example, when discussing the impact of different historical events in history class, students' attention to different perspectives such as social fairness and personal struggle reflects their value orientation; and their preference for astronomy and biology in the selection of topics for science projects reflects their interests.

[0055] The emotional resonance and cognitive guidance strategy generation module generates personalized emotional resonance and cognitive guidance strategies based on a two-dimensional analysis of students' cognition and emotions. It is used to provide teachers with strategic suggestions based on the characteristics of different students to achieve the best teaching results. For students who are emotionally sensitive to failure and have a weak cognitive foundation in a certain subject, teachers can first share some inspirational stories of celebrities overcoming similar difficulties in teaching, to trigger emotional resonance and enhance students' confidence, and then adopt a step-by-step, shallow-to-deep teaching method to guide them to gradually master knowledge.

[0056] The knowledge association map dynamic update module is used to construct and update the knowledge association map of each subject in real time. The knowledge association map is used to display the logical relationship between knowledge points, and based on the students' learning data, it marks each student's mastery of different knowledge points and learning paths. When it is found that students' learning difficulties in the mathematical function part affect their understanding of kinematic formulas in physics, the knowledge association map dynamic update module clearly presents this relationship through the constructed knowledge association map and provides teachers with teaching adjustment suggestions.

[0057] The module for building a personalized learning model accurately includes a multi-source data integration module, a feature extraction and analysis module, and a model building and verification module. The multi-source data integration module is used to collect students' learning behavior data, academic performance data, classroom interaction data, and self-assessment data. The learning behavior data includes students' operation records on various learning platforms; the academic performance data includes daily homework, periodic tests, and test scores, as well as the scores of each knowledge point and the distribution of answering time; the classroom interaction data is used to record students' speech content, question frequency, activeness in participating in group discussions, and interaction methods with classmates and teachers; the self-assessment data includes regularly collecting students' self-assessment questionnaires on their own learning status, interests, and goals;

[0058] The feature extraction and analysis module is used to extract and analyze students' cognitive abilities, learning styles, emotional attitudes, and interest preferences. Cognitive abilities are based on academic performance and learning behavior data to analyze students' memory, logical reasoning, and spatial imagination. Learning styles are determined based on students' preferences for learning materials and classroom interaction methods. Visual learners may rely more on visual materials such as charts and images. Auditory learners have a high acceptance of audio explanations. Kinesthetic learners are active in practical operations and group activities. Emotional attitudes are used to combine classroom interactions with self-assessment data to determine emotional factors such as students' learning motivation, self-confidence, and anxiety levels. Interest preferences are based on students' self-selected extended learning content, community activities they participate in, and topics they focus on in discussions. Students' cognitive abilities are scored using the formula:

[0059]

[0060] Sc is the student’s comprehensive cognitive ability score; is the student's score on the i-th cognitive ability dimension; n is the number of cognitive ability dimensions; w i is the weight of the i-th cognitive ability dimension. Cognitive ability is divided into excellent: 85-100 points; good: 70-84 points; medium: 60-69 points; poor: below 60 points;

[0061] The model building and verification module is based on the neural network model. It uses the extracted features as input to build a learning model unique to each student. The learning model is used to predict students' performance in future learning tasks, such as the accuracy of completing specific assignments and the time required to learn new knowledge, providing a basis for adjusting teaching content and progress. With the continuous input of new data, the learning model adjusts parameters in real time to achieve dynamic updates. When students take new exams or complete new projects, the model re-evaluates students' cognitive abilities and learning status based on new academic performance data, so that the model always fits the students' actual situation. The learning model is regularly verified through actual teaching results, and the predicted results of the learning model are compared with the students' actual learning performance. If the prediction deviation is large, the cause is analyzed and the model parameters and feature selection methods are adjusted to ensure the accuracy and reliability of the model. For example, if the model predicts that students will not have difficulty learning a certain knowledge point, but there are many errors in the actual assignments, the relevant feature analysis and model building logic need to be re-examined.

[0062] The decision-making module builds a personalized learning model for each student and provides teachers with interactive feedback on teaching adjustment strategies based on the personalized learning model.

[0063] The decision-making module includes the human-computer collaborative teaching decision-making module, the interdisciplinary integration strategy formulation module, and the teaching plan optimization module. The human-computer collaborative teaching decision-making module includes the suggestion diversification module, the teacher-led optimization module, and the real-time communication feedback module. The suggestion diversification module provides teachers with a variety of teaching adjustment strategy suggestions based on the personalized learning model constructed for students. For students with weak knowledge mastery and insufficient learning motivation, a gamification teaching method is adopted to integrate knowledge points into interesting games to stimulate students' learning interest. For students with strong abstract thinking ability but poor practical operation ability, experimental courses and practical projects are added to allow students to deepen their knowledge understanding through hands-on operations.

[0064] The teacher-led optimization module comprehensively evaluates and optimizes the suggestions provided by the suggestion diversification module based on the teacher's rich teaching experience, in-depth understanding of individual students and teaching goals. The teacher can adjust the suggestions based on the overall atmosphere of the class and the interpersonal relationships between students. For example, when the suggestion diversification module recommends using group competitions to improve students' learning enthusiasm, but the teacher considers that some students in the class are under great competitive pressure, the competition format is adjusted to a cooperative project, which not only achieves the purpose of stimulating students' motivation, but also avoids putting too much pressure on students.

[0065] The real-time communication feedback module is used to maintain real-time communication between teachers and the human-computer collaborative teaching decision-making module during the teaching decision-making process. Teachers can raise questions about the suggestions provided by the human-computer collaborative teaching decision-making module, and the human-computer collaborative teaching decision-making module will immediately provide explanations and additional explanations to promote further data analysis and provide more accurate suggestions for subsequent teaching decisions. When teachers try to suggest stratified teaching as recommended by the diversity module, they find that some middle-level students lose confidence in the learning process. After giving feedback to the human-computer collaborative teaching decision-making module, the human-computer collaborative teaching decision-making module provides teachers with new suggestions for adjusting teaching methods by re-analyzing the data of these students.

[0066] The interdisciplinary integration strategy formulation module analyzes the explicit connections between knowledge points of different disciplines based on the constructed knowledge association map and explores the potential deep-level connections. Based on this, an interdisciplinary integration teaching strategy is formulated to enable students to understand the occurrence and development of knowledge from multiple perspectives. When analyzing history and geography, it is found that the political and economic development of a specific historical period is closely related to the geographical environment at that time. Through an in-depth understanding of these connections, an interdisciplinary integration teaching strategy is formulated to enable students to understand the occurrence and development of historical events from multiple perspectives, and at the same time master the influence of geographical knowledge in the historical process; with complex problems and hot topics in real life as the theme, interdisciplinary integration teaching projects are designed to enable students to learn to use multidisciplinary knowledge to solve problems. For example, with the theme of sustainable urban development, the knowledge of urban planning and environmental resource protection in geography, the lessons learned from urban development changes in history, and the methods of data analysis and model construction in mathematics are integrated, so that students can learn to use multidisciplinary knowledge to solve practical problems by completing the project, and cultivate students' comprehensive thinking ability and innovation ability; at the same time, an interdisciplinary teacher communication and collaboration platform is built, on which teachers from different disciplines can share teaching experience, discuss integrated teaching strategies, and jointly design teaching content and evaluation methods for students;

[0067] The teaching plan optimization module is used to track students' learning status in real time. It dynamically adjusts the teaching plan according to students' immediate feedback in class, homework completion status and periodic test results. If it is monitored during the classroom explanation that most students have difficulty understanding a certain knowledge point, the module can analyze students' facial expressions and speech content to determine the students' confusion points and immediately remind teachers to adjust the teaching rhythm, add relevant cases and explanations to help students understand. Based on students' personalized learning models and learning progress, it formulates long-term learning path plans for students, covering the learning sequence and depth of knowledge in various subjects, as well as students' interest development and future career directions. For emergencies in the teaching process, including students' poor learning status due to special events and schools' temporary adjustment of teaching arrangements, it quickly analyzes the impact of events on students' learning and provides corresponding response decisions. When the school shortens the teaching time by one week due to special activities, it is recommended that teachers give priority to the teaching of key knowledge points and adjust teaching methods to improve learning efficiency according to the teaching progress of each subject and students' learning situation, such as adopting online and offline hybrid teaching, and using spare time to provide students with supplementary learning resources to ensure the realization of teaching goals. Based on the autoregressive integral moving average model, it predicts students' learning progress. Assume that the time series y t Indicates that students

[0068] The learning indicator at time t, the model expression is:

[0069] Θ(B)(1-B) d y t =Φ(B)∈ t ;

[0070] Where B is the backward shift operator; Θ(B) is the autoregressive part; Φ(B) is the sliding average part; ∈ t is a white noise sequence; d is the difference order; based on this, the autoregressive order p and the sliding average order q are determined, and the learning progress of future students is predicted by fitting the historical learning data.

[0071] The teaching content updating module re-arranges and optimizes the teaching content of teachers based on the decisions generated by the decision generation module;

[0072] The teaching content update module automatically rearranges and optimizes the teaching content based on teaching decisions, selects appropriate multimedia materials such as pictures, videos, and audios from the teaching resource library, organically integrates them with the original teaching content, and generates new teaching courseware. When explaining historical events, the system automatically inserts relevant historical video materials and expert interpretation audio to enhance the interest and attractiveness of teaching. In order to improve students' learning interest and knowledge application ability, situational teaching content is generated according to teaching objectives and students' actual conditions. For example, in English teaching, real life scenes are created, such as ordering food in a restaurant and waiting at an airport, so that students can practice dialogue in the context and improve their practical language application ability. Students are encouraged to participate in the creation and sharing of teaching content, and their excellent works and unique insights are integrated into teaching resources. For example, in composition teaching, excellent compositions of students are selected and displayed to the whole class, and analyzed and commented on, so that students can learn writing skills from their peers' works. In science classes, students' experimental reports and research results are organized into cases for other students to learn and refer to.

[0073] Embodiment 2:

[0074] A method for dynamically adjusting teaching content based on educational assistance includes the following steps:

[0075] S1: Multi-source data collection and analysis: collect students’ learning behavior data, academic performance data, classroom interaction data, and self-assessment data in the classroom, and use machine learning algorithms to deeply analyze these data to extract students’ cognitive ability characteristics, learning style characteristics, emotional attitude characteristics, and interest preference characteristics;

[0076] S2: Personalized learning model construction. Analyze the collected features and use the neural network model to build a unique learning model for each student to predict the student's future learning performance. Update and optimize in real time with new data input. At the same time, regularly verify the model through actual teaching results and calibrate deviations.

[0077] S3: Human-computer collaborative teaching decision-making, based on students' characteristic data and personalized learning models, provides teachers with a variety of teaching adjustment strategy suggestions. The teaching adjustment strategy suggestions include theoretical basis, expected effects and potential challenges. Teachers evaluate and optimize the suggestions based on their own teaching experience, understanding of individual students and teaching goals, and adjust the strategies according to the actual situation of the class. During the teaching decision-making process, teachers can ask questions about the suggestions; teachers can also provide feedback on problems found in practice, which is used to prompt them to further analyze and provide more accurate suggestions;

[0078] S4: Formulate interdisciplinary integration strategies, use knowledge association maps to deeply explore the explicit and potential connections between knowledge points in different disciplines, design interdisciplinary integration teaching projects based on real-life problems and hot topics, build an interdisciplinary teacher communication and collaboration platform, promote teachers from different disciplines to jointly design teaching content and evaluation methods, and achieve complementary disciplinary advantages;

[0079] S5: Dynamic teaching plan optimization, real-time tracking of students' classroom expressions, questions, participation, homework, and test results, based on which the teaching rhythm and content are dynamically adjusted. At the same time, based on students' personalized learning models and progress, long-term learning path plans are formulated, taking into account interests and career directions. In the face of teaching emergencies, the impact can be quickly analyzed and response decisions can be provided to ensure the achievement of teaching goals.

[0080] The above shows and describes the basic principles and main features of the present invention and the advantages of the present invention. It should be understood by those skilled in the art that the present invention is not limited to the above embodiments. The above embodiments and descriptions are only for explaining the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, which fall within the scope of the present invention to be protected. The scope of protection of the present invention is defined by the attached claims and their equivalents.

Claims

1. A teaching content dynamic adjustment system based on educational assistance, comprising a global perception module, an intelligent analysis module, a decision-making generation module, and a teaching content update module, characterized in that: The global perception module is used to capture students' behavioral details, environmental data and learning trajectories in the classroom, and to judge the teaching content based on the captured data, while understanding students' learning behavior; The intelligent analysis module is used to capture the behavior of students in the classroom and the environmental data of the classroom, analyze the current teaching effect based on the captured data, and integrate the external learning platform to grasp the students' learning; The decision generation module builds a unique personalized learning model for each student, and provides teachers with interactive feedback on teaching adjustment strategy suggestions based on the personalized learning model; The teaching content updating module rearranges and optimizes the teacher's teaching content based on the decision generated by the decision generating module.

2. The teaching content dynamic adjustment system based on educational assistance according to claim 1 is characterized by: The global perception module includes a behavior detail capture module, an environmental data integration module, and a learning platform deep integration module. The behavior detail capture module is based on the arrangement of high-definition cameras and precise sound pickup equipment in the classroom, and is used to record students' obvious behaviors in the classroom, including but not limited to speaking, raising hands to answer questions, and also to capture students' subtle behaviors in the classroom, including but not limited to the direction of eye focus, pauses and hesitations when writing. When the intelligent analysis module detects that most students' eyes are wandering when explaining a certain knowledge point, it means that the students' attention is distracted, and it is judged that the effect of explaining the knowledge point is not good at this time; The environmental data integration module collects classroom environmental data in all directions based on sensors, including temperature, humidity, light intensity, noise decibels, and analyzes it in combination with teaching time and course type; The learning platform deep integration module is used to deeply integrate with various online learning platforms, electronic textbooks, and homework systems to collect students' learning trajectories on different platforms, including homework completion status, test scores, annotation content on electronic textbooks, and download preferences for learning materials, so as to fully understand students' learning behavior.

3. The teaching content dynamic adjustment system based on educational assistance according to claim 1 is characterized by: The intelligent analysis module includes a cognitive and emotional dual-dimensional analysis module, a knowledge association map dynamic update module, and a personalized learning model precision construction module, wherein the cognitive and emotional dual-dimensional analysis module includes a multimodal data fusion analysis module, a knowledge transfer and cognitive network construction module, an emotional fluctuation prediction and context association module, a value and interest tendency mining module, and an emotional resonance and cognitive guidance strategy generation module. The multimodal data fusion analysis module is used to fuse and analyze text data, voice, expression, and body movements. When students answer questions, their voice intonation, speaking speed, facial expressions, and body movements are combined to more comprehensively judge and analyze their mastery of knowledge and their current emotional state; The knowledge transfer and cognitive network construction module is used to deeply analyze the students' knowledge transfer between different subjects and different knowledge points, and based on this, build a unique cognitive network for each student to demonstrate their mastery of each knowledge point and the strength of the association between knowledge points and the students' thinking path when applying knowledge. It is used to discover the weak links and potential advantages in the students' cognitive structure and provide an accurate basis for personalized learning.

4. The teaching content dynamic adjustment system based on educational assistance according to claim 3 is characterized by: The emotion fluctuation prediction and situation association module dynamically tracks and predicts the students' emotional state based on time series analysis and situation awareness technology, and adjusts the teaching method according to the tracking and prediction structure; The value and interest mining module mines students' potential values ​​and interest tendencies based on their opinions expressed in class discussions, homework and projects, and their topic preferences. It combines these mining results with teaching content to make teaching more targeted and attractive, and is used to stimulate students' intrinsic learning motivation. The emotional resonance and cognitive guidance strategy generation module generates personalized emotional resonance and cognitive guidance strategies based on a dual-dimensional analysis of students' cognition and emotions, and is used to provide teachers with strategic suggestions based on the characteristics of different students to achieve the best teaching results.

5. The teaching content dynamic adjustment system based on educational assistance according to claim 3 is characterized by: The knowledge association map dynamic update module is used to construct and update the knowledge association map of each subject in real time. The knowledge association map is used to display the logical relationship between knowledge points and mark each student's mastery of different knowledge points and learning path based on the students' learning data.

6. The teaching content dynamic adjustment system based on educational assistance according to claim 3 is characterized by: The personalized learning model precision construction module includes a multi-source data integration module, a feature extraction and analysis module, and a model construction and verification module. The multi-source data integration module is used to collect students' learning behavior data, academic performance data, classroom interaction data, and self-assessment data, wherein the learning behavior data includes students' operation records on various learning platforms; the academic performance data includes daily homework, periodic tests, and test scores, as well as the scores of each knowledge point and the distribution of answering time; the classroom interaction data is used to record the content of students' speeches in class, the frequency of questions, the activeness of participating in group discussions, and the interaction methods with classmates and teachers; the self-assessment data includes regularly collecting students' self-assessment questionnaires on their own learning status, interests, and goals; The feature extraction and analysis module is used to extract and analyze students' cognitive abilities, learning styles, emotional attitudes, and interest preferences. Cognitive abilities are based on academic performance and learning behavior data to analyze students' memory abilities, logical reasoning abilities, and spatial imagination abilities. Learning styles are determined based on students' preferences for learning materials and classroom interaction methods. Emotional attitude is used to combine classroom interaction and self-assessment data to judge students' emotional factors such as learning motivation, self-confidence, and anxiety level; interest preference is used to explore students' interest preferences from the extended learning content they choose, the community activities they participate in, and the topics they pay attention to in discussions, and score students' cognitive abilities through the formula: Sc is the student's comprehensive cognitive ability score; Sc i is the student's score on the i-th cognitive ability dimension; n is the number of cognitive ability dimensions; w i is the weight of the i-th cognitive ability dimension, and the cognitive ability is divided into standards according to the scores, which are divided into excellent: 85-100 points; Good: 70-84 points; Medium: 60-69 points; Poor: below 60 points; The model building and verification module is based on a neural network model, and uses the extracted features as input to build a learning model unique to each student. The learning model is used to predict the student's performance in future learning tasks. As new data is continuously input, the learning model adjusts parameters in real time to achieve dynamic updates. The learning model is regularly verified through actual teaching results, and the learning model prediction results are compared with the students' actual learning performance. If the prediction deviation is large, the cause is analyzed and the model parameters and feature selection methods are adjusted to ensure the accuracy and reliability of the model.

7. The teaching content dynamic adjustment system based on educational assistance according to claim 6 is characterized by: The decision generation module includes a human-computer collaborative teaching decision module, an interdisciplinary integration strategy formulation module, and a teaching plan optimization module. The human-computer collaborative teaching decision module includes a suggestion diversification module, a teacher-led optimization module, and a real-time communication feedback module. The suggestion diversification module provides teachers with a variety of teaching adjustment strategy suggestions based on the personalized learning model constructed for students. The teacher-led optimization module comprehensively evaluates and optimizes the suggestions provided by the suggestion diversification module based on the teacher's rich teaching experience, in-depth understanding of individual students and teaching goals. The teacher can adjust the suggestions based on the overall atmosphere of the class and the interpersonal relationships between students; The real-time communication feedback module is used to maintain real-time communication between the teacher and the human-computer collaborative teaching decision-making module during the teaching decision-making process. The teacher can raise questions about the suggestions provided by the human-computer collaborative teaching decision-making module, and the human-computer collaborative teaching decision-making module will immediately provide explanations and additional explanations to promote further data analysis and provide more accurate suggestions for subsequent teaching decisions.

8. The teaching content dynamic adjustment system based on educational assistance according to claim 1 is characterized by: The interdisciplinary integration strategy formulation module analyzes the explicit connections between knowledge points of different disciplines based on the constructed knowledge association map and explores the potential deep-level connections, and formulates interdisciplinary integration teaching strategies based on this, so as to enable students to understand the occurrence and development of knowledge from multiple perspectives; Design interdisciplinary teaching projects with themes based on complex problems and hot topics in real life, so that students can learn to use multidisciplinary knowledge to solve problems. At the same time, build an interdisciplinary teacher communication and collaboration platform, where teachers from different disciplines can share teaching experiences, discuss integrated teaching strategies, and jointly design teaching content and evaluation methods for students. The teaching plan optimization module is used to track students' learning status in real time, and dynamically adjust the teaching plan according to students' immediate feedback in class, homework completion status and periodic test results; based on students' personalized learning model and learning progress, a long-term learning path plan is formulated for students, covering the learning sequence and depth of knowledge in various subjects, as well as students' interest development and future career direction; for emergencies in the teaching process, including students' poor learning status due to special events and the school's temporary adjustment of teaching arrangements, the impact of the event on students' learning is quickly analyzed, and corresponding response decisions are provided. Based on the autoregressive integrated sliding average model, the learning progress of students is predicted. Assume that the time series y t represents the student’s learning index at time t, and the model expression is: I(B)(1-B) d y t =Φ(B)∈ t ; Where B is the backward shift operator; Θ(B) is the autoregressive part; Φ(B) is the sliding average part; ∈ t is a white noise sequence; d is the difference order; based on this, the autoregressive order p and the sliding average order q are determined, and the learning progress of future students is predicted by fitting the historical learning data.

9. The teaching content dynamic adjustment system based on educational assistance according to claim 1 is characterized by: The teaching content update module automatically rearranges and optimizes the teaching content based on teaching decisions, selects appropriate multimedia materials such as pictures, videos, and audios from the teaching resource library, organically integrates them with the original teaching content, and generates new teaching courseware.

10. A method for dynamically adjusting teaching content based on educational assistance, using a system for dynamically adjusting teaching content based on educational assistance as described in claims 1-9, characterized in that: The steps include: S1: Multi-source data collection and analysis: collect students’ learning behavior data, academic performance data, classroom interaction data, and self-assessment data in the classroom, and use machine learning algorithms to deeply analyze these data to extract students’ cognitive ability characteristics, learning style characteristics, emotional attitude characteristics, and interest preference characteristics; S2: Personalized learning model construction. Analyze the collected features and use the neural network model to build a unique learning model for each student to predict the student's future learning performance. Update and optimize in real time with new data input. At the same time, regularly verify the model through actual teaching results and calibrate deviations. S3: Human-computer collaborative teaching decision-making, based on students' characteristic data and personalized learning models, provides teachers with a variety of teaching adjustment strategy suggestions. The teaching adjustment strategy suggestions include theoretical basis, expected effects and potential challenges. Teachers evaluate and optimize the suggestions based on their own teaching experience, understanding of individual students and teaching goals, and adjust the strategies according to the actual situation of the class. During the teaching decision-making process, teachers can ask questions about the suggestions; teachers can also provide feedback on problems found in practice, which is used to prompt them to further analyze and provide more accurate suggestions; S4: Formulate interdisciplinary integration strategies, use knowledge association maps to deeply explore the explicit and potential connections between knowledge points in different disciplines, design interdisciplinary integration teaching projects based on real-life problems and hot topics, build an interdisciplinary teacher communication and collaboration platform, promote teachers from different disciplines to jointly design teaching content and evaluation methods, and achieve complementary disciplinary advantages; S5: Dynamic teaching plan optimization, real-time tracking of students' classroom expressions, questions, participation, homework, and test results, based on which the teaching rhythm and content are dynamically adjusted. At the same time, based on students' personalized learning models and progress, long-term learning path plans are formulated, taking into account interests and career directions. In the face of teaching emergencies, the impact can be quickly analyzed and response decisions can be provided to ensure the achievement of teaching goals.

Images