A Smart Classroom Control Method and System Based on Multimodal Cognitive State Assessment

Abstract

This invention discloses a smart classroom control method and system based on multimodal cognitive state assessment, relating to the field of teaching technology. The key technical points include the following steps: collecting multi-dimensional perception data of the smart classroom; wherein the multi-dimensional perception data includes student cognitive behavior data and teaching process state data; performing feature analysis on student cognitive behavior data to obtain cognitive features; continuously tracking and analyzing student cognitive feedback data to obtain individual cognitive fit and group cognitive synergy; verifying the cognitive synergy recovery state of the smart classroom based on individual cognitive fit and group cognitive synergy; and judging the cognitive synergy recovery state to determine whether classroom control is needed. The effect is to improve the utilization efficiency of teaching resources and reduce the consumption of ineffective teaching.

Description

Technical Field

[0001] This invention relates to the field of teaching technology, and more specifically, to a smart classroom control method and system based on multimodal cognitive state assessment. Background Technology

[0002] In current smart classroom teaching practices, traditional classroom control models generally have significant shortcomings. Most classrooms rely on teachers' subjective experience to judge the teaching status, only paying attention to the classroom performance of some students and failing to comprehensively cover the cognitive behavior details of all students. For example, teachers cannot simultaneously grasp each student's response speed, the true accuracy rate of knowledge point mastery, or accurately capture real-time changes in group interaction and participation, resulting in biased and lagging judgments of the classroom cognitive status. Traditional control lacks a dynamic closed-loop optimization mechanism, often lacking continuous effect verification and iteration after a single adjustment. For example, after adjusting the teaching pace, it is impossible to quantitatively evaluate the actual effect of the adjustment on students' cognitive improvement, and it is also difficult to update the control strategy in a timely manner based on students' subsequent feedback. As a result, the teaching pace and resource allocation are often out of sync with students' real-time cognitive needs. Summary of the Invention

[0003] In view of the shortcomings of existing technologies, the purpose of this invention is to provide a smart classroom control method and system based on multimodal cognitive state assessment.

[0004] To achieve the above objectives, the present invention provides the following technical solution: A smart classroom control method based on multimodal cognitive state assessment, comprising the following steps: Collect multi-dimensional perception data for smart classrooms; wherein, the multi-dimensional perception data includes student cognitive behavior data and teaching process status data; Cognitive features are obtained by analyzing student cognitive behavior data, and group adaptation features are obtained by analyzing teaching process state data. The cognitive collaboration status of the smart classroom is determined based on the cognitive features and group adaptation features. Determine whether the teaching pace and resource allocation mode need to be adjusted based on the cognitive synergy status, and collect student cognitive feedback data and teaching dynamic adjustment data; The regulation effect coefficient is obtained by analyzing student cognitive feedback data and teaching dynamic adjustment data. The regulation effect coefficient and student cognitive feedback data are combined to generate a smart classroom regulation plan. Continuous tracking and analysis of student cognitive feedback data yields individual cognitive fit and group cognitive synergy. The cognitive synergy recovery status of the smart classroom is then verified based on individual cognitive fit and group cognitive synergy. Assess the state of cognitive coordination recovery to determine whether classroom intervention is necessary.

[0005] Preferably, the cognitive features are obtained by performing feature analysis on students' cognitive behavior data, specifically including the following steps: The student cognitive behavior data includes classroom response time, accuracy rate of knowledge point mastery, and quality score of interactive questioning. Set cognitive behavior threshold ranges and duration of abnormal features; If any cognitive behavior data is outside the corresponding cognitive behavior threshold range and the duration exceeds the duration of feature abnormality, then the feature corresponding to that data is marked as a cognitively weak feature. If all cognitive behavior data are within the corresponding cognitive behavior threshold range, or the duration does not exceed the duration of the characteristic abnormality, then the temporal change trajectory of each data is determined, and characteristic indicators such as data improvement rate, stable fluctuation range and efficient duration are extracted based on the temporal change trajectory. Set the characteristic indicator evaluation thresholds, which include the threshold range for the rate of improvement, the threshold range for the fluctuation range, and the threshold for the duration of high efficiency. If all characteristic indicators meet the characteristic indicator evaluation threshold, the cognitive characteristic is judged as a high-quality cognitive characteristic; if at least one characteristic indicator does not meet the characteristic indicator evaluation threshold, the cognitive characteristic is judged as a cognitive characteristic that needs to be improved. Among them, high-quality cognitive features and cognitive features that need improvement are combined to form cognitive features.

[0006] Preferably, the group-fitting features are obtained by performing feature analysis on the teaching process state data, specifically including the following steps: The teaching process status data includes interactive participation rate, collaborative completion efficiency, and teaching content acceptance score. The rate of change of data was obtained based on the data from each teaching process and the corresponding monitoring period. Set a group adaptation threshold range, which includes the standard threshold range for each data and the threshold range for the duration of anomalies, the data change type, and the corresponding change trend rate. The teaching process data and the rate of change of the data are compared with the group fit threshold range. If all the teaching process data and the corresponding rate of change of the data meet the group fit threshold range, the group fit feature is judged to be a good fit feature. If at least one teaching process data point or its corresponding rate of change does not meet the group fit threshold range, then the group fit feature is determined to be a fit feature to be optimized.

[0007] Preferably, determining whether to adjust the teaching pace and resource allocation mode based on the cognitive coordination status, and collecting student cognitive feedback data and teaching dynamic adjustment data, specifically includes the following steps: Maintain the current teaching pace and resource allocation mode of the smart classroom when the cognitive collaboration state is in a collaborative adaptation state; When cognitive collaboration is in a state of mismatch, adjust the teaching pace and resource allocation mode, and collect student cognitive feedback data and teaching dynamic adjustment data.

[0008] Preferably, the control effect coefficient is obtained by analyzing the dynamic adjustment data of teaching and the cognitive feedback data of students, specifically including the following steps: The data on dynamic adjustments to teaching include the magnitude of adjustments to the teaching pace, the proportion of resource allocation adjustments, and the duration of the adjustments. Student cognitive feedback data includes the extent of cognitive improvement, satisfaction rating, and fit feedback level; Parameters for evaluating the control effect are generated based on teaching dynamic adjustment data and student cognitive feedback data. The regulation effect evaluation parameters are input into the regulation effect evaluation model to obtain the regulation effect coefficient.

[0009] Preferably, the smart classroom control plan is generated by combining the control effect coefficient and student cognitive feedback data, specifically including the following steps: The adjustable and optimized dimensions of the smart classroom are obtained; wherein, the adjustable and optimized dimensions include personalized learning task allocation, tiered teaching content design, interactive mode optimization, and auxiliary resource supply. If the regulation effect coefficient is less than or equal to the preset effect coefficient threshold, the current regulation is judged to be inefficient. Priority is given to optimizing the interactive mode and the supply of auxiliary resources. A regulation plan is generated by combining the shortcomings in the student cognitive feedback data. If the regulation effect coefficient is greater than the preset effect coefficient threshold, the current regulation is judged to be an efficient regulation. The regulation priority is divided according to the individual differences in the student cognitive feedback data. The regulation implementation intensity corresponding to each priority is determined in combination with the regulation effect coefficient. The regulation priority and the optimization measures corresponding to the corresponding regulation implementation intensity are integrated into the regulation plan.

[0010] Preferably, the regulation priority is determined based on individual differences in student cognitive feedback data, specifically including the following steps: Set at least one priority assessment interval, and each priority assessment interval corresponds to a control priority. Individual differences in student cognitive feedback data include differences in cognitive level, types of learning ability weaknesses, and urgency of adaptation needs. The weight coefficients of the differential characteristics are obtained based on the scope of influence and the urgency of improvement of each differential characteristic. Compare the differential feature weight coefficients with the priority evaluation intervals; The control priority corresponding to the priority evaluation interval where the differential feature weight coefficient is located is marked as the control priority of that differential feature.

[0011] Preferably, the intensity of regulation implementation corresponding to each priority level is determined by combining the regulation effect coefficient, specifically including the following steps: Obtain the standard benchmark parameters for the control of the smart classroom, and input the standard benchmark parameters into the control effect evaluation model to obtain the standard effect coefficient; Obtain the number of levels of regulatory priority, and divide the intervals based on the number of levels and the standard effect coefficient to obtain the effect coefficient gradient interval; The corresponding gradient interval of the effect coefficient is obtained by comparing the control effect coefficient with the effect coefficient gradient interval. The intensity of regulation implementation corresponding to each regulation priority is determined based on the gradient range of the effect coefficient.

[0012] Preferably, determining whether classroom intervention is necessary after assessing the state of cognitive coordination recovery includes the following steps: When the cognitive collaboration recovery state is in a fully collaborative state, the original teaching rhythm and resource allocation mode of the smart classroom are restored. When the cognitive synergy recovery state is in a partially synergistic state, the control plan is updated based on student cognitive feedback data and teaching dynamic adjustment data.

[0013] A smart classroom control system based on multimodal cognitive state assessment includes: Data Acquisition Module: Collects multi-dimensional sensory data for the smart classroom; wherein, the multi-dimensional sensory data includes student cognitive behavior data and teaching process status data; Analysis module: Performs feature analysis on student cognitive behavior data to obtain cognitive features, performs feature analysis on teaching process state data to obtain group adaptation features, and determines the cognitive collaboration status of the smart classroom based on cognitive features and group adaptation features; The first judgment module: determines whether the teaching pace and resource allocation mode need to be adjusted based on the cognitive coordination status, and collects student cognitive feedback data and teaching dynamic adjustment data; The first analysis module analyzes student cognitive feedback data and teaching dynamic adjustment data to obtain the regulation effect coefficient, and combines the regulation effect coefficient and student cognitive feedback data to generate a smart classroom regulation plan. The second analysis module continuously tracks and analyzes student cognitive feedback data to obtain individual cognitive fit and group cognitive synergy, and verifies the cognitive synergy recovery status of the smart classroom based on individual cognitive fit and group cognitive synergy. The second judgment module assesses the state of cognitive coordination recovery to determine whether classroom intervention is necessary.

[0014] Compared with the prior art, the present invention has the following beneficial effects: This invention collects student cognitive behavior data and teaching process status data, covering both individual student cognitive performance such as answer responses and knowledge point mastery, and group status such as classroom interaction and collaboration, avoiding the one-sidedness of single data. It ensures that the teaching pace and resources are always matched to the students' real-time status. It addresses the differentiated needs of students through individual cognitive fit, while ensuring overall classroom teaching efficiency through group cognitive synergy. It adjusts the supply of supplementary resources to address the cognitive weaknesses of some students, and optimizes overall participation through interactive modes, avoiding the problems of over-focusing on individuals and disrupting the classroom pace, or focusing only on the group and ignoring student differences. This ensures that students of different cognitive levels receive appropriate learning support in the classroom. It improves the efficiency of teaching resource utilization and reduces the consumption of ineffective teaching. Through the practical application of the effect coefficient evaluation measures, it can identify efficient control directions, reduce unnecessary resource adjustment costs, and allow teaching resources to more accurately serve students' cognitive improvement. Attached Figure Description

[0015] Figure 1 This is a schematic diagram illustrating the steps of a smart classroom control method based on multimodal cognitive state assessment, as provided in an embodiment of the present invention. Figure 2 This is a schematic diagram of a smart classroom control system based on multimodal cognitive state assessment, provided as an embodiment of the present invention. Detailed Implementation

[0016] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0017] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0018] Secondly, the term "an embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places throughout this specification does not necessarily refer to the same embodiment, nor is it a single embodiment or an embodiment selectively excluded from other embodiments.

[0019] Reference Figures 1-2 As shown.

[0020] The embodiments further illustrate the smart classroom control method and system based on multimodal cognitive state assessment proposed in this invention.

[0021] A smart classroom control method based on multimodal cognitive state assessment, comprising the following steps: Collect multi-dimensional perception data for smart classrooms; among which, multi-dimensional perception data includes student cognitive behavior data and teaching process status data; Cognitive features are obtained by analyzing student cognitive behavior data, and group adaptation features are obtained by analyzing teaching process state data. The cognitive collaboration status of the smart classroom is determined based on the cognitive features and group adaptation features. Determine whether the teaching pace and resource allocation mode need to be adjusted based on the cognitive synergy status, and collect student cognitive feedback data and teaching dynamic adjustment data; The regulation effect coefficient is obtained by analyzing student cognitive feedback data and teaching dynamic adjustment data. The regulation effect coefficient and student cognitive feedback data are combined to generate a smart classroom regulation plan. Continuous tracking and analysis of student cognitive feedback data yields individual cognitive fit and group cognitive synergy. The cognitive synergy recovery status of the smart classroom is then verified based on individual cognitive fit and group cognitive synergy. Assess the state of cognitive coordination recovery to determine whether classroom intervention is necessary.

[0022] The cognitive features are obtained by performing feature analysis on students' cognitive and behavioral data, specifically including the following steps: Student cognitive behavior data includes classroom response time, accuracy rate of knowledge mastery, and quality score of interactive questioning; Set cognitive behavior threshold ranges and duration of abnormal features; If any cognitive behavior data is outside the corresponding cognitive behavior threshold range and the duration exceeds the duration of feature abnormality, then the feature corresponding to that data is marked as a cognitively weak feature. If all cognitive behavior data are within the corresponding cognitive behavior threshold range, or the duration does not exceed the duration of the characteristic abnormality, then the temporal change trajectory of each data is determined, and characteristic indicators such as data improvement rate, stable fluctuation range and efficient duration are extracted based on the temporal change trajectory. Set the characteristic indicator evaluation thresholds, which include the threshold range for the rate of improvement, the threshold range for the fluctuation range, and the threshold for the duration of high efficiency. If all characteristic indicators meet the characteristic indicator evaluation threshold, the cognitive characteristic is judged as a high-quality cognitive characteristic; if at least one characteristic indicator does not meet the characteristic indicator evaluation threshold, the cognitive characteristic is judged as a cognitive characteristic that needs to be improved. Among them, high-quality cognitive features and cognitive features that need improvement are combined to form cognitive features.

[0023] Student cognitive behavior data includes classroom response time, accuracy rate of knowledge point mastery, and interactive question quality score. These three data points reflect students' classroom cognitive status from different dimensions. For example, classroom response time reflects the speed at which students understand the content of the questions, the accuracy rate of knowledge point mastery directly corresponds to the degree of knowledge absorption, and the interactive question quality score reflects the depth of students' thinking engagement.

[0024] Pre-set threshold ranges for cognitive behavior and durations of abnormal characteristics. For specific indicators, the threshold range for classroom response time is set to 10 to 30 seconds, and the duration of abnormal characteristics is 3 consecutive questions; the threshold range for knowledge point mastery accuracy is set to 80% to 100%, and the duration of abnormal characteristics is 2 consecutive knowledge point modules; the threshold range for interactive question quality score is set to 7 to 10 points, and the duration of abnormal characteristics is 2 consecutive questions.

[0025] If any cognitive behavior data point is outside its corresponding threshold range, and the duration of this state exceeds the preset duration of abnormal feature, the feature corresponding to that data point is marked as a cognitive weakness feature. For example, if a student's response time for answering three consecutive questions exceeds 30 seconds, then the feature corresponding to the student's classroom response time is judged as a cognitive weakness feature; if a student's correct mastery rate for two consecutive knowledge modules is below 80%, the feature corresponding to their knowledge module mastery rate is marked as a cognitive weakness feature.

[0026] If all cognitive behavioral data are within the corresponding threshold range, or if the duration of data exceeding the range does not reach the characteristic abnormal duration, then the temporal change trajectory of each data point is determined, and the data improvement rate, stable fluctuation range, and efficient duration are extracted from it. Taking the knowledge point mastery accuracy rate as an example, the temporal change trajectory can be continuous data with an accuracy rate of 80% for module 1, 85% for module 2, and 90% for module 3. The data improvement rate = (current module accuracy rate - initial module accuracy rate) ÷ number of modules; the stable fluctuation range is the difference between the maximum and minimum values ​​among multiple consecutive data points. For example, if the classroom answer response time is 12 seconds, 15 seconds, 13 seconds, 14 seconds, and 16 seconds for 5 consecutive questions, its stable fluctuation range is 16 seconds - 12 seconds = 4 seconds; the efficient duration refers to the continuous duration of data within the efficient range. For example, if the interactive question quality score is maintained above 8 points 4 times consecutively, its efficient duration is 4 classroom interactions.

[0027] It is necessary to set thresholds for characteristic indicators, specifically including threshold ranges for improvement rate, fluctuation range, and efficient duration. For example, for the improvement rate of knowledge point mastery accuracy, the threshold range is set to be no less than 2% per module; for the stable fluctuation range of classroom answer response time, the threshold range is set to be no more than 5 seconds; and for the efficient duration of interactive question quality scoring, the threshold is set to be no less than 3 classroom interactions.

[0028] If all characteristic indicators meet the corresponding characteristic indicator evaluation thresholds, the cognitive characteristic is determined to be a high-quality cognitive characteristic. For example, if a student's rate of improvement in knowledge point mastery is 3.33% per module, the stable fluctuation range of classroom answering time is 4 seconds, and the efficient duration of interactive questioning quality scoring is 4 times, if at least one characteristic indicator does not meet the evaluation threshold, the cognitive characteristic is determined to be a cognitive characteristic that needs improvement. For example, if a student's stable fluctuation range of classroom answering time is 6 seconds, even if other indicators meet the requirements, their cognitive characteristic is determined to be a cognitive characteristic that needs improvement. The final cognitive characteristic is the result of a combination of high-quality cognitive characteristics and cognitive characteristics that need improvement.

[0029] The process of analyzing teaching process status data to obtain group-fit ​​features includes the following steps: The data on the teaching process status includes interactive participation rate, collaborative completion efficiency, and teaching content acceptance rating; The rate of change of data was obtained based on the data from each teaching process and the corresponding monitoring period. Set the group adaptation threshold range, which includes the standard threshold range for each data point, the duration of the anomaly, the data change type, and the corresponding change rate threshold range. The teaching process data and the rate of change of the data are compared with the group fit threshold range. If all the teaching process data and the corresponding rate of change of the data meet the group fit threshold range, the group fit feature is judged to be a good fit feature. If at least one teaching process data point or its corresponding rate of change does not meet the group fit threshold range, then the group fit feature is determined to be a fit feature to be optimized.

[0030] The data on the teaching process includes interaction participation rate, collaborative completion efficiency, and teaching content acceptance score. The interaction participation rate reflects the overall level of student participation in class, the collaborative completion efficiency reflects the progress of group cooperative tasks, and the teaching content acceptance score corresponds to the degree to which the student group adapts to the current teaching content.

[0031] The rate of change of data trend = (current teaching process data - initial teaching process data) ÷ monitoring period duration. Taking the interaction participation rate as an example, if the interaction participation rate is 60% in the initial period (10 minutes), 80% in the current period (20 minutes), and the monitoring period duration is 10 minutes, then the rate of change of data trend = (80% - 60%) ÷ 10 minutes = 2% per minute; taking the collaborative completion efficiency as an example, if the completion progress is 10% in the initial period (when the task starts), 40% in the current period (5 minutes after the task starts), and the monitoring period duration is 5 minutes, then the rate of change of data trend = (40% - 10%) ÷ 5 minutes = 6% per minute.

[0032] It is necessary to pre-set the group adaptation threshold range, the standard threshold range and abnormal duration period for each teaching process data, and the threshold range for the data change type and corresponding rate of change. For specific indicators, the standard threshold range for interaction participation rate is set to 70% to 90%, with an abnormal duration period of 15 consecutive minutes; the standard threshold range for collaborative completion efficiency is set to 50% to 80% every 20 minutes, with an abnormal duration period of one consecutive task module; and the standard threshold range for teaching content acceptance score is set to 7 to 9 points, with an abnormal duration period of two consecutive teaching segments. If the interaction participation rate shows an upward trend, the corresponding rate of change threshold range is 1% to 3% per minute; if the collaborative completion efficiency shows a stable trend, the corresponding rate of change threshold range is 5% to 7% per minute; and if the teaching content acceptance score shows a steady trend, the corresponding rate of change threshold range is 0.1 to 0.3 points per segment.

[0033] The teaching process data and the rate of change of the data are compared with the preset group fit threshold range. If all teaching process data are within the corresponding standard threshold range and the duration does not exceed the abnormal duration period, and the rate of change of each data also meets the threshold range of the corresponding change type, then the group fit feature is determined to be a good fit feature. For example, if the interaction participation rate is maintained at 75% (between 70% and 90%) and the continuous duration does not exceed 15 minutes, the rate of change is 2% per minute (between 1% and 3% per minute); if the collaborative completion efficiency is 60% every 20 minutes (between 50% and 80%) and the continuous duration does not exceed one task module, the rate of change is 6% per minute (between 5% and 7% per minute); if the teaching content acceptance score is 8 points (between 7 and 9 points) and the continuous duration does not exceed two teaching segments, the rate of change is 0.2 points per segment (between 0.1 and 0.3 points), then the group fit feature is a good fit feature.

[0034] If at least one piece of teaching process data is outside the corresponding standard threshold range, or its duration exceeds the abnormal duration period, or its data change rate does not conform to the threshold range of the corresponding change type, then the group fit characteristic is determined to be a fit characteristic that needs optimization. For example, if the interaction participation rate drops to 65% (exceeding the 70% to 90% range) and the continuous duration reaches 20 minutes (exceeding the abnormal duration period of 15 minutes), even if other data meet the threshold requirements, the group fit characteristic is determined to be a fit characteristic that needs optimization.

[0035] Based on the cognitive synergy status, determine whether the teaching pace and resource allocation mode need to be adjusted, and collect student cognitive feedback data and teaching dynamic adjustment data, specifically including the following steps: Maintain the current teaching pace and resource allocation mode of the smart classroom when the cognitive collaboration state is in a collaborative adaptation state; When cognitive collaboration is in a state of mismatch, adjust the teaching pace and resource allocation mode, and collect student cognitive feedback data and teaching dynamic adjustment data.

[0036] Firstly, this solution is based on the cognitive collaboration state of the smart classroom, which is mainly divided into two categories: collaborative adaptation and collaborative mismatch. Only when the cognitive collaboration state is in the collaborative adaptation state will the current teaching pace and resource allocation mode of the smart classroom be maintained. The teaching pace includes the current speed of explaining knowledge points and the proportion of class practice time. The resource allocation mode covers the proportion of various learning resources provided, such as the duration of courseware presentations and the supply of group discussion resources. If the current class's knowledge point explanation speed is one module every 15 minutes, the proportion of class practice time is 30%, the ratio of courseware presentations to interactive resources in the resource allocation is 6:4, and the cognitive collaboration state is collaborative adaptation, then these parameters of the teaching pace and resource allocation will remain unchanged to continue the current adapted classroom state.

[0037] When cognitive synergy is in a state of mismatch, adjustments to the teaching pace and resource allocation model are initiated, while simultaneously collecting student cognitive feedback data and dynamic teaching adjustment data. Adjusting the teaching pace involves changing the speed of knowledge point explanation, such as slowing down the pace of a module from 15 minutes to 20 minutes, or increasing the proportion of class practice time to 40%. Adjusting the resource allocation model can involve changing the resource allocation ratio, such as adjusting the ratio of courseware presentations to interactive resources to 4:6, increasing the supply of interactive resources. During these adjustments, it is necessary to collect student cognitive feedback data and dynamic teaching adjustment data. Student cognitive feedback data includes the extent of cognitive improvement, while dynamic teaching adjustment data covers the adjustment range of the teaching pace and the resource allocation adjustment ratio.

[0038] The control effect coefficient is obtained by analyzing the dynamic adjustment data of teaching and the cognitive feedback data of students. The specific steps include: The data on dynamic adjustments to teaching include the magnitude of adjustments to the teaching pace, the proportion of resource allocation adjustments, and the duration of the adjustments. Student cognitive feedback data includes the extent of cognitive improvement, satisfaction rating, and fit feedback level; Parameters for evaluating the control effect are generated based on teaching dynamic adjustment data and student cognitive feedback data. The regulation effect evaluation parameters are input into the regulation effect evaluation model to obtain the regulation effect coefficient.

[0039] The data on dynamic adjustments to teaching includes the magnitude of the adjustment to the teaching pace, the proportion of resource allocation adjustments, and the duration of the adjustments. The magnitude of the teaching pace adjustment refers to the degree of change in the teaching pace before and after the adjustment. For example, if the original pace of explaining knowledge points was 15 minutes per module, and the adjusted pace is 20 minutes per module, the magnitude of the adjustment is (20-15)÷15≈33.33%. The proportion of resource allocation adjustments refers to the change in the percentage of various resources allocated. For example, if the original proportion of courseware resources was 60%, and the adjusted proportion is 40%, the proportion of the adjustment is 40%-60%=-20%. The duration of the adjustments refers to the duration of the control measures. For example, if the pace and resource adjustments last for 25 minutes.

[0040] Student cognitive feedback data includes cognitive improvement magnitude, satisfaction rating, and fit feedback level. Cognitive improvement magnitude refers to the degree of change in students' cognitive level after intervention. For example, if the accuracy rate of knowledge point mastery was 75% before intervention and 85% after intervention, the improvement magnitude is 85% - 75% = 10%. Satisfaction rating is the student's subjective evaluation score of the intervention measures, usually expressed on a scale of 1 to 10. Fit feedback level is a graded evaluation of the degree of matching between the intervention measures and students' needs, such as high fit, medium fit, and low fit.

[0041] Based on these two types of data, parameters for evaluating the effectiveness of regulation are generated. These parameters are a structured integration of dynamic adjustment data in teaching and cognitive feedback data in students. For example, the adjustment range of teaching pace (33.33%), resource allocation adjustment ratio (-20%), and adjustment implementation duration (25 minutes) are matched with the cognitive improvement of 10%, satisfaction score of 8 points, and high level of adaptability feedback.

[0042] The generated evaluation parameters for the control effect are input into the control effect evaluation model to obtain the control effect coefficient. The control effect evaluation model calculates each parameter according to preset weight rules. For example, the weight of the adjustment range of teaching rhythm is 0.2, the weight of the adjustment ratio of resource allocation is 0.2, the weight of the adjustment implementation time is 0.1, the weight of the cognitive improvement range is 0.3, the weight of the satisfaction score is 0.1, and the weight of the adaptability feedback level is 0.1 (the adaptability feedback level needs to be converted into a quantitative score first, for example, high adaptability corresponds to 9 points). The control effect coefficient = adjustment range of teaching rhythm × 0.2 + adjustment ratio of resource allocation × 0.2 + adjustment implementation time × 0.1 + cognitive improvement range × 0.3 + satisfaction score × 0.1 + quantitative value of adaptability feedback level × 0.1. Substituting the data, the control effect coefficient is obtained as follows: 33.33% × 0.2 + (-20%) × 0.2 + 25 × 0.1 + 10% × 0.3 + 8 × 0.1 + 9 × 0.1 ≈ 4.2567.

[0043] A smart classroom control plan is generated by combining the control effect coefficient and student cognitive feedback data, specifically including the following steps: The adjustable and optimizable dimensions of the smart classroom are obtained; these include personalized learning task allocation, tiered teaching content design, interactive mode optimization, and auxiliary resource supply. If the regulation effect coefficient is less than or equal to the preset effect coefficient threshold, the current regulation is judged to be inefficient. Priority is given to optimizing the interactive mode and the supply of auxiliary resources. A regulation plan is generated by combining the shortcomings in the student cognitive feedback data. If the regulation effect coefficient is greater than the preset effect coefficient threshold, the current regulation is judged to be an efficient regulation. The regulation priority is divided according to the individual differences in the student cognitive feedback data. The regulation implementation intensity corresponding to each priority is determined in combination with the regulation effect coefficient. The regulation priority and the optimization measures corresponding to the corresponding regulation implementation intensity are integrated into the regulation plan.

[0044] First, it's necessary to identify the adjustable and optimizeable dimensions of a smart classroom, including personalized learning task allocation, tiered instructional content design, interactive mode optimization, and supplementary resource provision. Personalized learning task allocation refers to assigning differentiated tasks based on students' individual cognitive levels; for example, assigning basic reinforcement tasks to students with weak cognitive abilities and extension tasks to students with strong cognitive abilities. Tiered instructional content design involves dividing instructional content into different difficulty levels to match students with different cognitive levels. Interactive mode optimization involves adjusting the form and frequency of classroom interaction, such as increasing the proportion of group discussions or Q&A sessions. Supplementary resource provision involves providing learning materials tailored to students' needs, such as providing specialized video explanations for students with weak knowledge points.

[0045] Based on the comparison between the control effect coefficient and the preset effect coefficient threshold, the first scenario is that the control effect coefficient is less than or equal to the preset effect coefficient threshold. In this case, the current control is determined to be inefficient. Priority is given to optimizing the interaction mode and the supply of auxiliary resources, and a control plan is generated by combining the shortcomings identified in the student cognitive feedback data. If the preset effect coefficient threshold is 5, and the current control effect coefficient is 4, and the shortcomings in the student cognitive feedback data are low interactive participation and insufficient understanding of knowledge points, then the interaction mode optimization can be adjusted to extend the original 10-minute group discussion to 15 minutes, and the auxiliary resource supply can be increased by adding diagrams of the corresponding knowledge points. Combining these two optimization measures forms the final control plan.

[0046] The second scenario is when the control effect coefficient exceeds the preset effect coefficient threshold, in which case the current control is deemed highly effective. Control priorities are determined based on individual differences in student cognitive feedback data. These individual differences include the degree of difference in cognitive level, the type of learning ability weakness, and the urgency of the adaptation needs. For example, students with high cognitive level differences will have higher priority in personalized learning task allocation; students whose learning ability weaknesses are concentrated in logical reasoning will have higher priority in the design of differentiated teaching content. The control implementation intensity is determined by combining the control effect coefficient: Control implementation intensity = Control effect coefficient ÷ Preset effect coefficient threshold × Priority base intensity. For example, if the preset effect coefficient threshold is 5, the current control effect coefficient is 6, and the base intensity of a certain priority is 3, then the control implementation intensity for that priority = 6 ÷ 5 × 3 = 3.6.

[0047] Based on individual differences in student cognitive feedback data, the priority of intervention is determined, which includes the following steps: Set at least one priority assessment interval, and each priority assessment interval corresponds to a control priority. Individual differences in student cognitive feedback data include differences in cognitive level, types of learning ability weaknesses, and urgency of adaptation needs. The weight coefficients of the differential characteristics are obtained based on the scope of influence and the urgency of improvement of each differential characteristic. Compare the differential feature weight coefficients with the priority evaluation intervals; The control priority corresponding to the priority evaluation interval where the differential feature weight coefficient is located is marked as the control priority of that differential feature.

[0048] At least one priority evaluation interval is preset, and each interval corresponds to a specific control priority. For example, three priority evaluation intervals can be set: interval one corresponds to high control priority, interval two corresponds to medium control priority, and interval three corresponds to low control priority. The value range of interval one is 0.7 to 1.0, the value range of interval two is 0.4 to 0.6, and the value range of interval three is 0.1 to 0.3.

[0049] Identify the individual differences in student cognitive feedback data, specifically including the degree of difference in cognitive level, the types of learning ability deficiencies, and the urgency of adaptation needs. The degree of difference in cognitive level refers to the disparity in cognitive levels among different individuals within a student group. For example, the highest accuracy rate of knowledge point mastery within a class might be 95%, while the lowest might be 60%, reflecting the dispersion of the group's cognitive level. The types of learning ability deficiencies refer to the categories of abilities commonly found in students during the learning process; for example, many students have weak logical reasoning skills. The urgency of adaptation needs refers to the urgency of students' need for certain types of learning support; for example, some students urgently need basic reinforcement guidance on specific knowledge points.

[0050] The weighting coefficient for the difference feature is calculated as follows: Impact range percentage × 0.6 + Improvement urgency score × 0.4. Here, the impact range percentage represents the proportion of students covered by the difference feature out of the total student population, and the improvement urgency score is expressed as a value from 0 to 1. For example, if the impact range percentage for cognitive level difference is 0.8 and the improvement urgency score is 0.9, the weighting coefficient is 0.8 × 0.6 + 0.9 × 0.4 = 0.84. If the impact range percentage for learning ability weakness is 0.6 and the improvement urgency score is 0.7, the weighting coefficient is 0.6 × 0.6 + 0.7 × 0.4 = 0.36 + 0.28 = 0.64. If the impact range percentage for the urgency of adaptation needs is 0.5 and the improvement urgency score is 0.8, the weighting coefficient is 0.5 × 0.6 + 0.8 × 0.4 = 0.3 + 0.32 = 0.62.

[0051] The weight coefficients of each difference feature are compared with preset priority evaluation intervals. The control priority corresponding to the interval in which the weight coefficient falls is marked as the control priority of that difference feature. For example, the weight coefficient of cognitive level difference is 0.84, which is in interval one, and the corresponding control priority is high; the weight coefficient of learning ability deficiency type is 0.64, which is in interval two, and the corresponding control priority is medium; the weight coefficient of adaptation need urgency is 0.62, which is in interval two, and the corresponding control priority is also medium.

[0052] The determination of the intensity of regulation implementation for each priority level is based on the regulation effectiveness coefficient, and specifically includes the following steps: Obtain the standard benchmark parameters for the control of the smart classroom, and input the standard benchmark parameters into the control effect evaluation model to obtain the standard effect coefficient; Obtain the number of levels of regulatory priority, and divide the intervals based on the number of levels and the standard effect coefficient to obtain the effect coefficient gradient interval; The corresponding gradient interval of the effect coefficient is obtained by comparing the control effect coefficient with the effect coefficient gradient interval. The intensity of regulation implementation corresponding to each regulation priority is determined based on the gradient range of the effect coefficient.

[0053] First, it's necessary to obtain the baseline parameters for smart classroom control. These parameters serve as benchmark reference values ​​for classroom control, such as a standard adjustment range of 20% for the teaching pace, 15% for resource allocation, and a 20-minute adjustment duration. These baseline parameters are then input into the control effectiveness evaluation model to obtain the standard effectiveness coefficient.

[0054] Obtain the number of priority levels for regulation. For example, if the current regulation priority is divided into three levels: high, medium, and low, the number of levels is 3. Based on this number of levels and the standard effect coefficient, divide the data into intervals to obtain the effect coefficient gradient intervals. The gradient interval length = standard effect coefficient ÷ number of levels. Substituting the data, we get the gradient interval length = 5 ÷ 3 ≈ 1.67. Therefore, the resulting effect coefficient gradient intervals are: Interval 1 corresponds to 0 to 1.67, Interval 2 corresponds to 1.67 to 3.34, and Interval 3 corresponds to 3.34 to 5.

[0055] The actual control effect coefficient is compared with the gradient interval of the effect coefficient to determine its corresponding gradient interval. For example, if the current actual control effect coefficient is 4, the comparison shows that it falls within interval three.

[0056] The intensity of regulation implementation corresponding to each regulation priority is determined based on the gradient range of the effect coefficient. Generally, the higher the value of the gradient range, the greater the intensity of regulation implementation. For example, the intensity of regulation implementation is 1 for interval 1, 2 for interval 2, and 3 for interval 3. If the regulation priority of a certain differential feature is high and its corresponding effect coefficient is in interval 3, then the intensity of regulation implementation corresponding to this priority is 3, which is specifically reflected in practical measures such as increasing the coverage ratio of personalized learning task allocation to 80%.

[0057] After assessing the state of cognitive synergy recovery, it is determined whether classroom intervention is necessary. This process includes the following steps: When the cognitive collaboration recovery state is in a fully collaborative state, the original teaching rhythm and resource allocation mode of the smart classroom are restored. When the cognitive synergy recovery state is in a partially synergistic state, the control plan is updated based on student cognitive feedback data and teaching dynamic adjustment data.

[0058] The cognitive synergy recovery state is clearly defined as including two categories: fully synergistic state and partially synergistic state. Individual cognitive fit reflects the degree of matching between the individual student's cognitive state and the teaching requirements, while group cognitive synergy reflects the consistency and fit of the overall cognitive state of the class.

[0059] When cognitive collaboration is fully restored, the original teaching rhythm and resource allocation model of the smart classroom are restored. The original teaching rhythm includes the speed of knowledge point explanation and the proportion of class session time before adjustment. For example, the original rhythm was to explain one knowledge point module every 15 minutes, with class practice accounting for 30% of the time. The original resource allocation model includes the proportion of various teaching resources, such as the ratio of courseware presentation to interactive resources being 6:4. If the speed of knowledge point explanation was adjusted to one module every 20 minutes due to collaboration mismatch, it will be adjusted back to one module every 15 minutes after the full collaboration is restored, and the resource allocation ratio will be restored to the initial 6:4 model to return to an appropriate and regular classroom state.

[0060] When the cognitive synergy recovery state is at a partially synergistic level, the control plan is updated based on student cognitive feedback data and dynamic teaching adjustment data. Student cognitive feedback data includes the extent of cognitive improvement and satisfaction ratings, while dynamic teaching adjustment data includes the previous adjustment range of teaching pace and resource allocation adjustment parameters. For example, if student cognitive feedback data shows that some students' correct mastery rate of knowledge points is still below 80%, and dynamic teaching adjustment data shows that the optimization range of the interaction mode is 20%, then updating the control plan can increase the optimization range of the interaction mode to 30%, while simultaneously increasing the supply of auxiliary resources for these students.

[0061] A smart classroom control system based on multimodal cognitive state assessment includes: Data Acquisition Module: Collects multi-dimensional sensory data for the smart classroom; this multi-dimensional sensory data includes student cognitive behavior data and teaching process status data. Analysis module: Performs feature analysis on student cognitive behavior data to obtain cognitive features, performs feature analysis on teaching process state data to obtain group adaptation features, and determines the cognitive collaboration status of the smart classroom based on cognitive features and group adaptation features; The first judgment module: determines whether the teaching pace and resource allocation mode need to be adjusted based on the cognitive coordination status, and collects student cognitive feedback data and teaching dynamic adjustment data; The first analysis module analyzes student cognitive feedback data and teaching dynamic adjustment data to obtain the regulation effect coefficient, and combines the regulation effect coefficient and student cognitive feedback data to generate a smart classroom regulation plan. The second analysis module continuously tracks and analyzes student cognitive feedback data to obtain individual cognitive fit and group cognitive synergy, and verifies the cognitive synergy recovery status of the smart classroom based on individual cognitive fit and group cognitive synergy. The second judgment module assesses the state of cognitive coordination recovery to determine whether classroom intervention is necessary.

[0062] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0063] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0064] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A smart classroom control method based on multimodal cognitive state assessment, characterized in that, The method includes the following steps: Collect multi-dimensional perception data for smart classrooms; wherein, the multi-dimensional perception data includes student cognitive behavior data and teaching process status data; Cognitive features are obtained by analyzing student cognitive behavior data, and group adaptation features are obtained by analyzing teaching process state data. The cognitive collaboration status of the smart classroom is determined based on the cognitive features and group adaptation features. Determine whether the teaching pace and resource allocation mode need to be adjusted based on the cognitive synergy status, and collect student cognitive feedback data and teaching dynamic adjustment data; The regulation effect coefficient is obtained by analyzing student cognitive feedback data and teaching dynamic adjustment data. The regulation effect coefficient and student cognitive feedback data are combined to generate a smart classroom regulation plan. Continuous tracking and analysis of student cognitive feedback data yields individual cognitive fit and group cognitive synergy. The cognitive synergy recovery status of the smart classroom is then verified based on individual cognitive fit and group cognitive synergy. Assess the state of cognitive coordination recovery to determine whether classroom intervention is necessary.

2. The smart classroom control method based on multimodal cognitive state assessment according to claim 1, characterized in that, The cognitive features are obtained by performing feature analysis on students' cognitive and behavioral data, specifically including the following steps: The student cognitive behavior data includes classroom response time, accuracy rate of knowledge point mastery, and quality score of interactive questioning. Set cognitive behavior threshold ranges and duration of abnormal features; If any cognitive behavior data is outside the corresponding cognitive behavior threshold range and the duration exceeds the duration of feature abnormality, then the feature corresponding to that data is marked as a cognitively weak feature. If all cognitive behavior data are within the corresponding cognitive behavior threshold range, or the duration does not exceed the duration of the characteristic abnormality, then the temporal change trajectory of each data is determined, and characteristic indicators such as data improvement rate, stable fluctuation range and efficient duration are extracted based on the temporal change trajectory. Set the characteristic indicator evaluation thresholds, which include the threshold range for the rate of improvement, the threshold range for the fluctuation range, and the threshold for the duration of high efficiency. If all characteristic indicators meet the characteristic indicator evaluation threshold, the cognitive characteristic is judged as a high-quality cognitive characteristic; if at least one characteristic indicator does not meet the characteristic indicator evaluation threshold, the cognitive characteristic is judged as a cognitive characteristic that needs to be improved. Among them, high-quality cognitive features and cognitive features that need improvement are combined to form cognitive features.

3. The smart classroom control method based on multimodal cognitive state assessment according to claim 2, characterized in that, The process of analyzing teaching process status data to obtain group-fit ​​features includes the following steps: The teaching process status data includes interactive participation rate, collaborative completion efficiency, and teaching content acceptance score. The rate of change of data was obtained based on the data from each teaching process and the corresponding monitoring period. Set a group adaptation threshold range, which includes the standard threshold range for each data and the threshold range for the duration of anomalies, the data change type, and the corresponding change trend rate. The teaching process data and the rate of change of the data are compared with the group fit threshold range. If all the teaching process data and the corresponding rate of change of the data meet the group fit threshold range, the group fit feature is judged to be a good fit feature. If at least one teaching process data point or its corresponding rate of change does not meet the group fit threshold range, then the group fit feature is determined to be a fit feature to be optimized.

4. The smart classroom control method based on multimodal cognitive state assessment according to claim 3, characterized in that, Based on the cognitive synergy status, determine whether the teaching pace and resource allocation mode need to be adjusted, and collect student cognitive feedback data and teaching dynamic adjustment data, specifically including the following steps: Maintain the current teaching pace and resource allocation mode of the smart classroom when the cognitive collaboration state is in a collaborative adaptation state; When cognitive collaboration is in a state of mismatch, adjust the teaching pace and resource allocation mode, and collect student cognitive feedback data and teaching dynamic adjustment data.

5. The smart classroom control method based on multimodal cognitive state assessment according to claim 4, characterized in that, The control effect coefficient is obtained by analyzing the dynamic adjustment data of teaching and the cognitive feedback data of students. The specific steps include: The data on dynamic adjustments to teaching include the magnitude of adjustments to the teaching pace, the proportion of resource allocation adjustments, and the duration of the adjustments. Student cognitive feedback data includes the extent of cognitive improvement, satisfaction rating, and fit feedback level; Parameters for evaluating the control effect are generated based on teaching dynamic adjustment data and student cognitive feedback data. The regulation effect evaluation parameters are input into the regulation effect evaluation model to obtain the regulation effect coefficient.

6. The smart classroom control method based on multimodal cognitive state assessment according to claim 5, characterized in that, A smart classroom control plan is generated by combining the control effect coefficient and student cognitive feedback data, specifically including the following steps: The adjustable and optimized dimensions of the smart classroom are obtained; wherein, the adjustable and optimized dimensions include personalized learning task allocation, tiered teaching content design, interactive mode optimization, and auxiliary resource supply. If the regulation effect coefficient is less than or equal to the preset effect coefficient threshold, the current regulation is judged to be inefficient. Priority is given to optimizing the interactive mode and the supply of auxiliary resources. A regulation plan is generated by combining the shortcomings in the student cognitive feedback data. If the regulation effect coefficient is greater than the preset effect coefficient threshold, the current regulation is judged to be an efficient regulation. The regulation priority is divided according to the individual differences in the student cognitive feedback data. The regulation implementation intensity corresponding to each priority is determined in combination with the regulation effect coefficient. The regulation priority and the optimization measures corresponding to the corresponding regulation implementation intensity are integrated into the regulation plan.

7. The smart classroom control method based on multimodal cognitive state assessment according to claim 6, characterized in that, Based on individual differences in student cognitive feedback data, the priority of intervention is determined, which includes the following steps: Set at least one priority assessment interval, and each priority assessment interval corresponds to a control priority. Individual differences in student cognitive feedback data include differences in cognitive level, types of learning ability weaknesses, and urgency of adaptation needs. The weight coefficients of the differential characteristics are obtained based on the scope of influence and the urgency of improvement of each differential characteristic. Compare the differential feature weight coefficients with the priority evaluation intervals; The control priority corresponding to the priority evaluation interval where the differential feature weight coefficient is located is marked as the control priority of that differential feature.

8. The smart classroom control method based on multimodal cognitive state assessment according to claim 7, characterized in that, The determination of the intensity of regulation implementation for each priority level is based on the regulation effectiveness coefficient, and specifically includes the following steps: Obtain the standard benchmark parameters for the control of the smart classroom, and input the standard benchmark parameters into the control effect evaluation model to obtain the standard effect coefficient; Obtain the number of levels of regulatory priority, and divide the intervals based on the number of levels and the standard effect coefficient to obtain the effect coefficient gradient interval; The corresponding gradient interval of the effect coefficient is obtained by comparing the control effect coefficient with the effect coefficient gradient interval. The intensity of regulation implementation corresponding to each regulation priority is determined based on the gradient range of the effect coefficient.

9. A smart classroom control method based on multimodal cognitive state assessment according to claim 8, characterized in that, After assessing the state of cognitive synergy recovery, it is determined whether classroom intervention is necessary. This process includes the following steps: When the cognitive collaboration recovery state is in a fully collaborative state, the original teaching rhythm and resource allocation mode of the smart classroom are restored. When the cognitive synergy recovery state is in a partially synergistic state, the control plan is updated based on student cognitive feedback data and teaching dynamic adjustment data.

10. A smart classroom control system based on multimodal cognitive state assessment, applied to the smart classroom control method based on multimodal cognitive state assessment as described in any one of claims 1 to 9, characterized in that, include: Data Acquisition Module: Collects multi-dimensional sensory data for the smart classroom; wherein, the multi-dimensional sensory data includes student cognitive behavior data and teaching process status data; Analysis module: Performs feature analysis on student cognitive behavior data to obtain cognitive features, performs feature analysis on teaching process state data to obtain group adaptation features, and determines the cognitive collaboration status of the smart classroom based on cognitive features and group adaptation features; The first judgment module: determines whether the teaching pace and resource allocation mode need to be adjusted based on the cognitive coordination status, and collects student cognitive feedback data and teaching dynamic adjustment data; The first analysis module analyzes student cognitive feedback data and teaching dynamic adjustment data to obtain the regulation effect coefficient, and combines the regulation effect coefficient and student cognitive feedback data to generate a smart classroom regulation plan. The second analysis module continuously tracks and analyzes student cognitive feedback data to obtain individual cognitive fit and group cognitive synergy, and verifies the cognitive synergy recovery status of the smart classroom based on individual cognitive fit and group cognitive synergy. The second judgment module assesses the state of cognitive coordination recovery to determine whether classroom intervention is necessary.

Images