A classroom teaching quality evaluation method, system and terminal based on a multi-person brain-computer interface

Abstract

The application belongs to the technical field of brain-computer interface, and relates to a classroom teaching quality evaluation method, system and terminal based on a multi-person brain-computer interface. The evaluation method comprises the following steps: in a pre-class stage, first, second and third evaluation parameters are calculated to obtain first and second grade thresholds; in a class stage, whether the first, second and third evaluation parameters of each student are greater than or equal to the corresponding first grade threshold is judged, if yes, A-level teaching quality is output, if not, whether the first, second and third evaluation parameters are greater than or equal to the corresponding second grade threshold is judged, if yes, B-level teaching quality is output, otherwise, C-level teaching quality is output; and the teaching quality grades corresponding to the three evaluation parameters are weighted and summed to obtain a final teaching quality grade. The application can effectively measure the explicit knowledge level and implicit cognitive level of students, reduce the interruption of traditional methods to teaching activities, avoid environmental noise and interference, enhance the accuracy of classroom teaching quality evaluation, and have stronger cross-scene generalization.

Description

Technical Field

[0001] This invention relates to the field of brain-computer interface technology, and in particular to a method, system and terminal for evaluating classroom teaching quality based on a multi-person brain-computer interface. Background Technology

[0002] Classroom assessment techniques (CATs) typically aim to improve student learning efficiency and classroom teaching effectiveness. They establish evaluation methods for a quantifiable learning outcome and conduct a series of simple, anonymous, non-grading formative classroom assessment activities during the teaching process. By reviewing assessment evidence and interpreting results, they provide feedback on learning status to teachers and students, enabling appropriate interventions. Traditional classroom assessment techniques include subjective scales (self-report scales, peer evaluation scales) and machine vision (facial expression recognition, head movement detection / eye tracking). Among these, subjective scales are considered the primary method of traditional classroom assessment due to their advantages such as intuitiveness, ease of use, low cost, and highly generalized information. However, subjective scales have the following main shortcomings: First, students are usually asked to fill out questionnaires after the teaching activities are completed, so the assessment results are delayed; second, the assessment results are based on students' subjective feedback and self-experience, which is affected by retrospective bias and social expectation bias, and is seriously affected by students' age, education level and comprehension ability; third, the use of scales (especially data collection) will interrupt teaching activities or occupy teaching time, creating a teaching load; fourth, repeated use of scales involves multiple completions of the items to be measured, and the test subjects will present fixed subjective report results, reducing the validity of repeated evaluation.

[0003] Image processing methods such as machine vision (facial expression recognition, head movement detection / eye tracking) identify students' learning emotions by monitoring involuntary facial muscle movements caused by changes in the emotional state of teachers or learners, or indirectly assess teaching status by detecting head movements, postures, and eye movements of teachers or learners to identify visual attention distribution and drowsiness levels. Compared to traditional subjective report scales, machine vision has the advantages of: ease of deployment, typically using classroom video surveillance systems as a hardware and software platform for secondary development; objective evaluation results, requiring no subjective evaluation or feedback from learners; and low teaching workload, without interrupting teaching activities. However, machine vision methods still have the following shortcomings:

[0004] (1) In a classroom environment with dense seating, the image signal-to-noise ratio is low, which will degrade the system performance, such as non-frontal facial images, changes in lighting, image occlusion, etc.

[0005] (2) Behavioral information is difficult to (directly) reflect cognitive state. It mainly focuses on evaluating drowsiness / awakeness, emotions, visual attention resource allocation, etc., and indirectly evaluates cognitive state such as participation and concentration.

[0006] Therefore, finding neurophysiological characteristics with high signal-to-noise ratio that are directly related to students' cognitive state and establishing a user-friendly real-time cognitive state monitoring method that does not interrupt teaching activities is a feasible way to achieve classroom teaching quality assessment.

[0007] To address the aforementioned issues, the use of brain-computer interface (BCI) technology for assessing classroom teaching quality (e.g., attention, learning efficiency) has garnered attention from educators. A BCI is a system that decodes brain activity into machine instructions without relying on normal output pathways composed of peripheral nerves and muscles, thereby enabling communication between the brain and the external world. For example, BCI-driven wheelchairs or exoskeletons can help patients with amyotrophic lateral sclerosis (ALS) or spinal cord injuries regain mobility. It can recognize specific brain signal patterns, involving five consecutive stages: signal acquisition, preprocessing, feature extraction, classification interface, and state feedback / control.

[0008] Based on the different user-oriented tasks (for feature formation) or control requirements of the application system, BCI can be divided into three types: active-BCI, reactive-BCI, and passive-BCI. These three types of systems respectively use the user's subjective interaction intention (active), the perception of external stimuli as required (reactive), and cognitive state (passive) to form / accompany neural activity as the front-end decoding features of brain-computer communication commands. Unlike the former two, the purpose of passive-BCI is not control, but to continuously monitor brain activity over a period of time and evaluate changes in the user's cognitive and emotional state. This is used as input for other adaptive systems to adjust human-computer interaction or task execution modes, thereby improving work efficiency. BCI provides many applications in the design of adaptive systems. These systems can adjust their behavior according to the user's continuous cognitive and emotional state without distracting the user's attention from the main task, thereby improving interaction quality and performance. Among them, the development of BCI systems using electroencephalography (EEG) signals is mainstream, as it has portability, high temporal resolution, and relatively low cost, and can be deployed in various applications, including human-computer interaction. EEG contains a wealth of physiological information and is an extremely complex bioelectrical signal, providing crucial analytical reference information for neuroscience, medicine, and other disciplines. EEG is a non-invasive method for acquiring brain neurophysiological activity by placing a signal sensor on the surface of the scalp to measure and record brain neural activity. Due to its millisecond-level temporal resolution and sensitivity to state fluctuations, it has significant advantages in state monitoring. For EEG-BCI systems, high signal-to-noise ratio EEG characteristics closely related to a specific cognitive state are a crucial initial factor determining its practical application, and recognition accuracy is the main indicator for evaluating the performance of passive BCI systems.

[0009] The main challenges of using brain-computer interfaces for classroom teaching quality assessment are:

[0010] (1) Existing technologies treat students and teachers as isolated individuals for state monitoring, making it difficult to measure changes in cognitive state at the interpersonal scale;

[0011] (2) The brain patterns related to cognitive state extracted by existing technologies may be affected by exogenous neural artifacts (electromyography, electrooculography) and task-irrelevant endogenous neural activity, leading to deterioration of BCI recognition performance;

[0012] (3) Existing technologies are usually developed in laboratory environments. Due to the differences between laboratory environments and classroom environments in terms of behavior control, environmental noise and interference levels, single-modal BCI systems built under laboratory conditions are difficult to extend to real classroom scenarios. Summary of the Invention

[0013] The purpose of this invention is to address the shortcomings of existing brain-computer interfaces for classroom teaching quality assessment, and to propose a method, system, and terminal for classroom teaching quality assessment based on a multi-person brain-computer interface.

[0014] To achieve the above objectives, the present invention adopts the following technical solution:

[0015] In a first aspect, the present invention provides a method for evaluating classroom teaching quality based on a multi-person brain-computer interface, comprising the following steps:

[0016] In the pre-class stage, the level thresholds of the first, second and third assessment parameters are calculated respectively; the level thresholds include at least level I level threshold and level II level threshold, and at least two level thresholds divide the classroom teaching quality into level A, level B and level C teaching quality from high to low; the first assessment parameter is the brain-brain coupling value of individual student-student group, the second assessment parameter is the brain-brain coupling value of individual student-teacher, and the third assessment parameter is the EEG-speech frequency following response value of individual student-teacher;

[0017] During the lesson, for each individual student, the first, second, and third assessment parameters are calculated. It is then determined whether each of these parameters is greater than or equal to the corresponding Level I threshold. If so, Level A teaching quality is output; otherwise, it is further determined whether each of these parameters is greater than or equal to the corresponding Level II threshold. If so, Level B teaching quality is output; otherwise, Level C teaching quality is output. For each individual student, the weighted sum of the teaching quality levels output for the first, second, and third assessment parameters is used to obtain the final teaching quality level.

[0018] As one possible implementation, in the after-class stage, level thresholds for the fourth and fifth assessment parameters are configured respectively. The level thresholds include at least a Level I threshold and a Level II threshold. At least two level thresholds divide the classroom teaching quality into Level A, Level B, and Level C teaching quality from high to low. The fourth assessment parameter is the student's individual memory efficiency, and the fifth assessment parameter is the student's individual subjective evaluation. The system calculates and determines whether the fourth assessment parameter is greater than or equal to its corresponding Level I threshold. If so, Level A teaching quality is output; otherwise, it further determines whether it is greater than or equal to its corresponding Level II threshold. If so, Level B teaching quality is output; otherwise, Level C teaching quality is output. The system also calculates and determines whether the fifth assessment parameter is greater than or equal to its corresponding Level I threshold. If so, Level A teaching quality is output; otherwise, it further determines whether it is greater than or equal to its corresponding Level II threshold. If so, Level B teaching quality is output; otherwise, Level C teaching quality is output.

[0019] For each individual student, the teaching quality level determined during the in-class stage and the teaching quality level determined after class stage are weighted and summed to obtain the final teaching quality level.

[0020] As one possible implementation, in the pre-class stage, the grade thresholds for the first, second, and third evaluation parameters are calculated using the following method:

[0021] Calculate the first, second, and third baseline parameters corresponding to the first, second, and third evaluation parameters respectively; where the first and second baseline parameters are the average brain-brain coupling of individual students and student groups in the pre-class stage, the average brain-brain coupling of individual students and teacher in a natural resting state without eye contact, and the average brain-brain coupling of individual students and teacher, respectively; the third baseline parameter is the average brain-frequency following response of individual students and neural activity frequency bands in the pre-class stage in a natural resting state within a preset time.

[0022] The Level I thresholds for the first, second, and third evaluation parameters are calculated as follows: and Level II threshold :

[0023]

[0024]

[0025] Among them, when hour, This represents the first baseline parameter corresponding to the first evaluation parameter, when... hour, This represents the second baseline parameter corresponding to the second evaluation parameter, when hour, This indicates the third baseline parameter corresponding to the third evaluation parameter.

[0026] As one possible implementation, cross-sample entropy is applied to measure brain-brain coupling values, which include brain-brain coupling values ​​between individual students and teachers, as well as brain-brain coupling values ​​between individual students. The samples are EEG time-series signals x and y obtained at the same time from the same leads and frequency band of individual students, teachers / other students. The entropy values ​​of EEG time-series signals x and y are calculated, specifically by sequentially performing signal reconstruction, vector spacing calculation, threshold comparison, probability solution, and entropy calculation on the two EEG time-series signals x and y to obtain the cross-sample entropy.

[0027] As one possible implementation, the first evaluation parameter, namely the brain-brain coupling value between individual students and the student group, is calculated as follows:

[0028] Calculate the cross-sample entropy value between the EEG signals of the current student and any other student in the same lead and frequency band. Obtain the set of cross-sample entropy values ​​of the number of students in the test group minus 1, which includes all pairs between the current student and other students.

[0029] Normalize the set of cross-sample entropy values;

[0030] The normalized cross-sample entropy values ​​are averaged over time and pairwise scales to obtain the average cross-sample entropy of the current student and the student group.

[0031] After obtaining the average cross-sample entropy within a predefined region of interest, a second average is performed to obtain the second average of the cross-sample entropy.

[0032] As one possible implementation, the second evaluation parameter, namely the student-teacher brain-brain coupling value, is calculated as follows:

[0033] Calculate the cross-sample entropy value between the EEG signals of the currently tested student and the teacher in the same lead and frequency band;

[0034] The time periods during which teachers and students engage in verbal and eye contact are defined as the time periods of interest. After obtaining the cross-sample entropy values ​​within the time periods of interest, the average values ​​are calculated to obtain the average cross-sample entropy between the current student and the teacher.

[0035] After obtaining the average cross-sample entropy within a predefined region of interest, a second average is performed to obtain the second average of the cross-sample entropy.

[0036] As one possible implementation, the third evaluation parameter, namely the student-teacher EEG-speech frequency following response value, is calculated as follows:

[0037] Extract the teacher's voice signal;

[0038] Multiple sub-band signals were obtained by filtering the EEG time-series signal of each student.

[0039] The instantaneous phase of the teacher's speech signal and multiple sub-band signals are obtained by Hilbert transform.

[0040] The student-teacher EEG-speech frequency following response value is measured by the phase lock value between the instantaneous phase of the teacher's speech signal and the instantaneous phase of multiple sub-band signals of the student.

[0041] As one possible implementation, the fourth evaluation parameter, namely the individual student's memory efficiency, can be obtained through the following method:

[0042]

[0043] in, This refers to the number of questions answered correctly in the post-test. This refers to the number of questions answered correctly in both the pre-test and post-test. This refers to the total number of in-class quiz questions, which is usually a constant.

[0044] Secondly, the present invention provides a classroom teaching quality assessment system based on a multi-person brain-computer interface, comprising:

[0045] The data acquisition module includes an EEG signal acquisition unit, used to collect EEG signals of students and teachers in a natural, resting state without eye contact before class, and to collect EEG time-series signals of students and teachers during class; a speech acquisition unit, used to collect speech signals of teachers during the lecture; and a behavior acquisition unit, used to present in-class tests and obtain behavioral data of students based on the test results; the EEG time-series signals, speech signals, and behavioral data are uploaded to a real-time data stream server and then sent to the data processing module after synchronization.

[0046] The data processing module performs feature extraction, pattern recognition, and fusion decision-making on EEG time-series signals, speech signals, and behavioral data during the lesson, and then outputs the teaching quality level.

[0047] It also includes an assessment and feedback module, which presents the teaching quality level of all students to teachers and the teaching quality level of individual students to students.

[0048] Thirdly, the present invention provides a terminal including a processor and a communication interface coupled to the processor, the processor being used to run computer programs or instructions to implement the classroom teaching quality assessment method based on a multi-person brain-computer interface provided in the first aspect.

[0049] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0050] 1. Compared with traditional single-person, single-modal BCI, the proposed solution of this invention adopts multi-person hybrid BCI technology, which regards the group in the interactive state as the monitoring subsystem, and can fully measure the interpersonal scale social interaction behavior and neural processes in the teaching process.

[0051] 2. The classroom teaching quality assessment method based on multi-person brain-computer interface proposed in this invention obtains brain-brain coupling from the neural activity between individuals, which can effectively suppress the influence of exogenous neural artifacts (electromyography, electrooculography) and task-irrelevant endogenous neural activity.

[0052] 3. The classroom teaching quality assessment method and system based on multi-person brain-computer interface proposed in this invention comprehensively considers individual EEG features, group EEG features, and behavioral features, and applies cross-sample entropy to measure brain-brain coupling value to improve the signal-to-noise ratio. Furthermore, the multimodal fusion strategy effectively avoids environmental noise and interference, enhances the accuracy of classroom teaching quality assessment, and has stronger cross-scenario generalization ability. Attached Figure Description

[0053] The accompanying drawings, which are included to provide a further understanding of the invention and form part of this invention, illustrate exemplary embodiments of the invention and are used to explain the invention, but do not constitute an undue limitation of the invention. In the drawings:

[0054] Figure 1 A flowchart of a classroom teaching quality assessment method based on a multi-person brain-computer interface provided in an embodiment of the present invention;

[0055] Figure 2 This is a flowchart illustrating the calculation of the first, second, and third baseline parameters corresponding to the first, second, and third evaluation parameters in an embodiment of the present invention.

[0056] Figure 3 This is a flowchart illustrating the process of obtaining teaching quality levels based on the fourth and fifth evaluation parameters in an embodiment of the present invention.

[0057] Figure 4 This is a schematic diagram of the classroom teaching quality assessment system based on a multi-person brain-computer interface in an embodiment of the present invention.

[0058] Figure Labels

[0059] 1-Data acquisition module, 10-EEG signal acquisition unit, 11-Voice acquisition unit, 12-Behavior acquisition unit, 2-Data processing module, 3-Evaluation and feedback module. Detailed Implementation

[0060] To facilitate a clear description of the technical solutions in the embodiments of the present invention, the terms "first" and "second" are used to distinguish identical or similar items with essentially the same function and effect. For example, the first threshold and the second threshold are merely used to distinguish different thresholds and do not limit their order. Those skilled in the art will understand that the terms "first" and "second" do not limit the quantity or execution order, and that the terms "first" and "second" are not necessarily different.

[0061] It should be noted that in this invention, the terms "exemplary" or "for example" are used to indicate examples, illustrations, or descriptions. Any embodiment or design described as "exemplary" or "for example" in this invention should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of terms such as "exemplary" or "for example" is intended to present the relevant concepts in a concrete manner.

[0062] In this invention, "at least one" refers to one or more, and "more than one" refers to two or more. "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, or B alone, where A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one" or similar expressions refer to any combination of these items, including any combination of singular or plural items. For example, "at least one of a, b, or c" can represent: a, b, c, a combination of a and b, a combination of a and c, a combination of b and c, or a, b, and c, where a, b, and c can be single or multiple.

[0063] Traditional brain-computer interface (BCI)-based classroom assessment methods often use single-person EEG data as input, focusing on the power of specific frequency components, the power ratio of different frequency band components, and cross-frequency coupling as features, emphasizing feedback on attention assessment. For example, parietal theta absolute power, central alpha absolute power, frontal theta / beta ratio, and theta-gamma coupling are used as electrophysiological markers for attention control and memory encoding. In a laboratory environment, power features are usually relatively stable, and the technical path for attention detection is well-defined, typically involving data acquisition, filtering, energy / energy ratio calculation, attention level discrimination based on prior thresholds, and feedback. However, these features lose interpersonal-scale behavioral and neural information, and their signal-to-noise ratio (SNR) can be significantly attenuated due to changes in the external environment (teacher's voice, room temperature), endogenous task-irrelevant neural activity, and exogenous neural artifacts (EMG, EEG), leading to deterioration in BCI recognition performance and limiting its application.

[0064] To address the shortcomings of existing technologies, this invention aims to provide a method, system, and terminal for evaluating classroom teaching quality based on a multi-person brain-computer interface. By collecting and quantifying EEG and behavioral data from teachers and students, and incorporating interpersonal-scale behavioral and neural information, the invention suppresses the influence of various noises and significantly improves evaluation performance.

[0065] In a first aspect, embodiments of the present invention provide a method for evaluating classroom teaching quality based on a multi-person brain-computer interface, see [link to relevant documentation]. Figure 1 It includes the following steps:

[0066] In the pre-class phase, the grade thresholds for the first, second, and third assessment parameters are calculated respectively;

[0067] As one possible implementation, in the pre-class stage, the grade thresholds for the first, second, and third evaluation parameters are calculated using the following method:

[0068] See Figure 2 The first, second, and third baseline parameters corresponding to the first, second, and third evaluation parameters are calculated respectively. Among them, the first and second baseline parameters are the average brain-brain coupling of the student group and the student group in the pre-class stage, the average brain-brain coupling of the student individual and the teacher in a natural resting state without eye contact, and the average brain-brain coupling of the student individual and the teacher.

[0069] As one possible implementation, cross-sample entropy is applied to measure brain-brain coupling values, which include brain-brain coupling values ​​between individual students and teachers, as well as brain-brain coupling values ​​between individual students. The samples are EEG time-series signals x and y obtained at the same time from the same leads and frequency band of individual students, teachers / other students. The entropy values ​​of EEG time-series signals x and y are calculated, specifically by sequentially performing signal reconstruction, vector spacing calculation, threshold comparison, probability solution, and entropy calculation on the two EEG time-series signals x and y to obtain the cross-sample entropy.

[0070] This invention employs the idea of ​​sample entropy to calculate cross-sample entropy to measure the similarity and complexity of two signals. Cross-sample entropy is an improved method of cross-approximate entropy (CAE), which is not direction-dependent and therefore has better relative consistency under various conditions.

[0071] As an example, the following method is used to reconstruct the signals from two EEG time series signals x and y sequentially:

[0072] Suppose there exists a one-dimensional time series obtained by sampling at equal time intervals. , Define parameters , .in, , represents the length of the metric vector. , representing a metric of "similarity", reconstructs the EEG time-series signal x into ,in, Reconstructing the EEG time-series signal y into ,in, .

[0073] The vector spacing is calculated using the following method:

[0074] Define vector and The distance between them is the Chebyshev distance, that is:

[0075]

[0076] in, , .

[0077] The following method is used for threshold comparison and probability calculation:

[0078] For a given ,statistics and Distance between of The number of, among which , recorded as In reconstructing the signal Calculate each vector with vector Distance between The probability of:

[0079]

[0080] in, .

[0081] In the embedding dimension is When calculating similarity tolerance Template matching probability below:

[0082]

[0083] Increase the embedding dimension to Repeat the above steps to calculate. .

[0084] The entropy value is calculated as follows:

[0085] For practical computing applications with limited dimensions In a 3D time series, the brain-brain coupling value between individual students and teachers is:

[0086]

[0087] Typically, the embedding dimension is taken. The similarity tolerance is 1 or 2. With an entropy value ranging from 0.1 to 0.25, the obtained entropy value exhibits reasonable statistical properties. This embodiment uses the embedding dimension. The similarity tolerance is 2. Let's take 0.2 as an example. Cross-entropy value representing each 2-second-long EEG signal segment (e.g., FP1-FP1, FP2-FP2) under one-to-one pairing of each electrode channel for subject x and y.

[0088] The brain-brain coupling value between individual students and the student population is obtained by averaging the complexity of all possible individual student-student pairings:

[0089]

[0090] Among the students He is one of the students in the lecture; the above formula represents a student. and students CSE between EEG signal pairs in the same channel and frequency band, For students Available logarithms ( ≤3), The proposed index represents the student population. The calculated CSE values ​​are normalized using the Fisher Z-transform and then averaged across trials, lectures, and coupled pairs for a more stable measure. Next, CSE values ​​are averaged across five predefined regions of interest: frontal, central, parietal, temporal, and occipital electrodes. Finally, data from question-specific periods determined based on lecture recordings are averaged, rather than averaging CSE values ​​over the entire duration of each lecture. Following this logic, the proposed index measures the overall similarity and complexity of neural activity between students and the population, or between students and teachers, in brain regions and frequency bands of interest, quantifying the coupling of neural activity between brain regions in a non-linear manner.

[0091] The grading thresholds include at least a Level I grading threshold and a Level II grading threshold. At least two grading thresholds divide classroom teaching quality into Level A, Level B, and Level C teaching quality from high to low. The first assessment parameter is the brain-brain coupling value between the individual student and the student group, the second assessment parameter is the brain-brain coupling value between the individual student and the teacher, and the third assessment parameter is the EEG-speech frequency follow-up response value between the individual student and the teacher.

[0092] As one possible implementation, the Level I thresholds for the first, second, and third evaluation parameters are calculated as follows: and Level II threshold :

[0093]

[0094]

[0095] Among them, when hour, This represents the first baseline parameter corresponding to the first evaluation parameter, when... hour, This represents the second baseline parameter corresponding to the second evaluation parameter, when hour, This indicates the third baseline parameter corresponding to the third evaluation parameter.

[0096] See Figure 1 During the lesson, for each individual student, the first, second, and third assessment parameters are calculated. It is then determined whether each of these parameters is greater than or equal to the corresponding Level I threshold. If so, Level A teaching quality is output; otherwise, it is further determined whether each of the first, second, and third parameters is greater than or equal to the corresponding Level II threshold. If so, Level B teaching quality is output; otherwise, Level C teaching quality is output. The determination results are shown in the following formula:

[0097] .

[0098] As one possible implementation, the first evaluation parameter, namely the brain-brain coupling value between individual students and the student group, is calculated as follows:

[0099] Calculate the cross-sample entropy value between the EEG signals of the current student and any other student in the same lead and frequency band. Obtain the set of cross-sample entropy values ​​of the number of students in the test group minus 1, which includes all pairs between the current student and other students.

[0100] Normalize the set of cross-sample entropy values;

[0101] The normalized cross-sample entropy values ​​are averaged over time and pairwise scales to obtain the average cross-sample entropy of the current student and the student group.

[0102] After obtaining the average cross-sample entropy within a predefined region of interest, a second average is performed to obtain the second average of the cross-sample entropy.

[0103] As one possible implementation, the second evaluation parameter, namely the student-teacher brain-brain coupling value, is calculated as follows:

[0104] Calculate the cross-sample entropy value between the EEG signals of the currently tested student and the teacher in the same lead and frequency band;

[0105] The time periods during which teachers and students engage in verbal and eye contact are defined as the time periods of interest. After obtaining the cross-sample entropy values ​​within the time periods of interest, the average values ​​are calculated to obtain the average cross-sample entropy between the current student and the teacher.

[0106] After obtaining the average cross-sample entropy within a predefined region of interest, a second average is performed to obtain the second average of the cross-sample entropy.

[0107] Driven by external rhythmic stimuli, the oscillatory activity within the brain gradually synchronizes with the rhythm of those stimuli; this phenomenon is known as external rhythmic synchronization of neural oscillations. When the external stimulus is a speech signal, the auditory cortex tracks the amplitude modulation of the input speech signal. This entrainment effect is mainly manifested in the waveform of low-frequency neural activity and the power envelope of high-frequency neural activity. This invention uses the phase locking value (PLV) between the speech signal amplitude envelope and the waveform of a specific frequency band of EEG signal to quantify the phase synchronization between EEG neural activity and speech stimulation, thereby reflecting the level of stimulus-brain coupling.

[0108] As one possible implementation, the third evaluation parameter, namely the student-teacher EEG-speech frequency following response value, is calculated as follows:

[0109] Extracting the teacher's voice signal;

[0110] Multiple sub-band signals were obtained by filtering the EEG time-series signal of each student.

[0111] The instantaneous phase of the teacher's speech signal and multiple sub-band signals are obtained by Hilbert transform.

[0112] The student-teacher EEG-speech frequency following response value is measured by the phase lock value between the instantaneous phase of the teacher's speech signal and the instantaneous phase of multiple sub-band signals of the student.

[0113] As an example, consider a one-dimensional time series of a speech signal. ,by Divide it into frames based on frame length. For each frame, the maximum amplitude is calculated as the amplitude envelope of that frame, i.e., for the ... The amplitude envelope of the frame sample is:

[0114]

[0115] in, For frame samples No. The amplitude of each sample point is used to connect the amplitude envelope of each frame to obtain the amplitude envelope of the speech signal:

[0116]

[0117] Since reducing the sampling rate of the speech signal leads to the loss of extracted amplitude envelope information, the envelope of the original speech signal with a sampling rate of 20kHz is extracted, and the envelope signal is downsampled to 500Hz to obtain... .

[0118] EEG filtering yields four sub-band signals. and voice envelope signal Analytical signals are calculated using Hilbert transform. :

[0119]

[0120]

[0121] in, It means Hilbert transform, This represents a real-valued signal, such as a filtered EEG sub-band signal. or voice envelope signal , It is a plural unit. This refers to amplitude. It refers to phase.

[0122] The PLV between the EEG signal and the speech envelope signal is defined as:

[0123]

[0124] in, For the number of trials, , and These represent the instantaneous phases of the EEG signal and the corresponding speech amplitude envelope at a certain frequency band. The PLV value ranges from 0 to 1. If the coupling level between the two signals is high, then... The fluctuation will be smaller, and the larger the PLV value, the smaller the fluctuation.

[0125] The third baseline parameter is the average EEG-frequency following response of individual students in a natural resting state within a preset time period during the pre-class phase. It is used to calculate the level threshold of the third assessment parameter, i.e.:

[0126]

[0127]

[0128]

[0129] in, This is the third baseline parameter. This refers to the Level I threshold of the third evaluation parameter. This is the Level II threshold for the third evaluation parameter.

[0130] For each individual student, the teaching quality level is obtained by weighted summing of the outputs of the first, second, and third assessment parameters.

[0131] See Figure 3 As one possible implementation, in the after-class stage, level thresholds for the fourth and fifth assessment parameters are configured respectively. These level thresholds include at least a Level I threshold and a Level II threshold. These at least two level thresholds divide classroom teaching quality from high to low into Level A, Level B, and Level C teaching quality. The fourth assessment parameter is the student's individual memory efficiency, and the fifth assessment parameter is the student's individual subjective evaluation. The system calculates and determines whether the fourth assessment parameter is greater than or equal to its corresponding Level I threshold. If so, Level A teaching quality is output; otherwise, it further determines whether it is greater than or equal to its corresponding Level II threshold. If so, Level B teaching quality is output; otherwise, Level C teaching quality is output. Similarly, the system calculates and determines whether the fifth assessment parameter is greater than or equal to its corresponding Level I threshold. If so, Level A teaching quality is output; otherwise, it further determines whether it is greater than or equal to its corresponding Level II threshold. If so, Level B teaching quality is output; otherwise, Level C teaching quality is output.

[0132] As one possible implementation, the fourth evaluation parameter, namely the individual student's memory efficiency, can be obtained through the following method:

[0133]

[0134] in, This refers to the number of questions answered correctly in the post-test. This refers to the number of questions answered correctly in both the pre-test and post-test. This refers to the total number of in-class quiz questions, which is usually a constant.

[0135] For the fourth evaluation parameter:

[0136]

[0137] in, This refers to the AB grade classification threshold value of evaluation parameter four. , This refers to the BC grade classification threshold value of assessment parameter four. .

[0138] Questions answered incorrectly in the pre-test but correctly in the post-test indicate that students performed effective memory retention, suggesting high working memory efficiency. Questions answered incorrectly in the post-test indicate that students failed to perform memory retention, suggesting low working memory efficiency. Questions answered correctly in both the pre-test and post-test are considered invalid samples because it's impossible to determine whether participants performed memory retention during stimulus presentation. This embodiment calculates memory efficiency based on all question responses, i.e., the proportion of samples with high working memory levels to the total number of valid samples, and converts this to a percentage as memory efficiency. This calculation method avoids evaluation errors caused by differences in prior knowledge levels between individuals, and the value range is between 0 and 100, with no abrupt changes.

[0139] As an example, the fifth assessment parameter is based on the 7-point Likert scale for students to evaluate their subjective experiences in terms of focus, knowledge acquisition, and participation. The degree description labels are set as follows: 1 None, 2 Rare, 3 Somewhat, 4 Moderate, 5 Many, 6 Quite a lot, 7 Extremely. For the fifth assessment parameter:

[0140]

[0141] in, This refers to the AB grade classification threshold value of evaluation parameter four. , This refers to the BC grade classification threshold value of assessment parameter four. .

[0142] For each individual student, the teaching quality level determined during the in-class stage and the teaching quality level determined after class stage are weighted and summed to obtain the final teaching quality level.

[0143] Secondly, embodiments of the present invention provide a classroom teaching quality assessment system based on a multi-person brain-computer interface, see [link to relevant documentation]. Figure 4 ,include:

[0144] The data acquisition module 1 includes an EEG signal acquisition unit 10, used to collect EEG signals of students and teachers in a natural, resting state without eye contact before class, and to collect EEG time-series signals of students and teachers during class; a speech acquisition unit 11, used to collect speech signals of teachers during the teaching process; and a behavior acquisition unit 12, used to present in-class tests and obtain behavioral data of students based on the test results; the EEG time-series signals, speech signals, and behavioral data are uploaded to a real-time data stream server and then sent to the data processing module 2 after synchronization.

[0145] As an example, EEG signals, speech signals, and behavioral signals generated by teachers' teaching and students' learning intentions are collected. After the three types of signals are subjected to analog signal processing such as denoising and amplification and A / D conversion, they are uploaded to a real-time data stream server to ensure time alignment of the multimodal data.

[0146] Data processing module 2 performs feature extraction, pattern recognition, and fusion decision-making on EEG time-series signals, speech signals, and behavioral data during the lesson, and then outputs the teaching quality level.

[0147] And the assessment and feedback module 3 is used to present the teaching quality level of all students to the teachers and the teaching quality level of each student to the students.

[0148] Thirdly, the present invention provides a terminal including a processor and a communication interface coupled to the processor, the processor being used to run computer programs or instructions to implement the classroom teaching quality assessment method based on a multi-person brain-computer interface provided in the first aspect.

[0149] Although the invention has been described herein in conjunction with various embodiments, those skilled in the art will understand and implement other variations of the disclosed embodiments by reviewing the accompanying drawings, the disclosure, and the description of the drawings, in carrying out the claimed invention. In this specification, the word "comprising" does not exclude other components or steps, and "a" or "an" does not exclude multiple components. A single processor or other unit can implement several of the functions listed in the specification. While certain measures are described in different embodiments, this does not mean that these measures cannot be combined to produce good results.

[0150] Although the invention has been described in conjunction with specific features and embodiments, it is obvious that various modifications and combinations can be made therein without departing from the spirit and scope of the invention. Accordingly, this specification and drawings are merely illustrative of the invention and are considered to cover any and all modifications, variations, combinations, or equivalents within the scope of the invention. Clearly, those skilled in the art can make various alterations and modifications to the invention without departing from its spirit and scope. Thus, if such modifications and modifications fall within the scope of the invention and its equivalents, the invention is also intended to include such modifications and modifications.

Claims

1. A method for evaluating classroom teaching quality based on multi-person brain-computer interface, characterized in that, Includes the following steps: In the pre-class stage, the level thresholds for the first, second, and third assessment parameters are calculated respectively. The level thresholds include at least a Level I threshold and a Level II threshold. At least two level thresholds divide the classroom teaching quality into Level A, Level B, and Level C teaching quality from high to low. The first assessment parameter is the brain-brain coupling value between the individual student and the student group, the second assessment parameter is the brain-brain coupling value between the individual student and the teacher, and the third assessment parameter is the EEG-speech frequency following response value between the individual student and the teacher. The third assessment parameter, namely the EEG-speech frequency following response value between the individual student and the teacher, is calculated by the following method: extracting the teacher's speech signal; filtering the EEG time series signal of each individual student to obtain multiple sub-frequency band signals; obtaining the instantaneous phase of the teacher's speech signal and the multiple sub-frequency band signals through Hilbert transform; measuring the EEG-speech frequency following response value between the individual student and the teacher through the phase lock value between the instantaneous phase of the teacher's speech signal and the instantaneous phase of the multiple sub-frequency band signals of the individual student. During the lesson, for each individual student, the first, second, and third assessment parameters are calculated. It is then determined whether each of these parameters is greater than or equal to the corresponding Level I threshold. If so, a Level A teaching quality is output; otherwise, it is further determined whether each of these parameters is greater than or equal to the corresponding Level II threshold. If so, a Level B teaching quality is output; otherwise, a Level C teaching quality is output. For each individual student, the weighted sum of the teaching quality levels corresponding to the first, second, and third assessment parameters is used to obtain the final teaching quality level. The first assessment parameter, namely the brain-brain coupling value between individual students and the student group, is calculated as follows: Calculate the cross-sample entropy value between the EEG signals of the current student and any other student in the same lead and frequency band. Obtain the set of cross-sample entropy values ​​of the number of students in the test group minus 1, which includes all pairs between the current student and other students. Normalize the set of cross-sample entropy values; The normalized cross-sample entropy values ​​are averaged on the time scale and the paired scale to obtain the average cross-sample entropy of the current student and the student group. The cross-sample entropy is used to measure the brain-brain coupling value, which includes the brain-brain coupling value between the student and the teacher and the brain-brain coupling value between students. The samples are EEG time-series signals x and y obtained at the same time from individual students and teachers / other students, with the same leads and frequency band. The entropy values ​​of the EEG time-series signals x and y are calculated, which specifically includes performing signal reconstruction, vector spacing calculation, threshold comparison, probability solution, and entropy calculation on the two EEG time-series signals x and y in sequence to obtain the cross-sample entropy. After obtaining the average cross-sample entropy within a predefined region of interest, a second average is performed to obtain the second average of the cross-sample entropy.

2. The classroom teaching quality assessment method based on multi-person brain-computer interface according to claim 1, characterized in that, In the after-class stage, the level thresholds for the fourth and fifth assessment parameters are configured respectively. The level thresholds include at least a Level I threshold and a Level II threshold. At least two level thresholds divide the classroom teaching quality into Level A, Level B, and Level C teaching quality from high to low. The fourth assessment parameter is the student's individual memory efficiency, and the fifth assessment parameter is the student's individual subjective evaluation. Calculate and determine whether the fourth assessment parameter is greater than or equal to its corresponding Level I threshold. If so, output Level A teaching quality; otherwise, further determine whether it is greater than or equal to its corresponding Level II threshold. If so, output Level B teaching quality; otherwise, output Level C teaching quality. Calculate and determine whether the fifth evaluation parameter is greater than or equal to its corresponding Level I threshold. If so, output Level A teaching quality; otherwise, further determine whether it is greater than or equal to its corresponding Level II threshold. If so, output Level B teaching quality; otherwise, output Level C teaching quality. For each individual student, the teaching quality level determined during the in-class stage and the teaching quality level determined after class stage are weighted and summed to obtain the final teaching quality level.

3. The classroom teaching quality assessment method based on multi-person brain-computer interface according to claim 1, characterized in that, In the pre-class phase, the grade thresholds for the first, second, and third assessment parameters are calculated using the following method: Calculate the first, second, and third baseline parameters corresponding to the first, second, and third evaluation parameters respectively; where the first and second baseline parameters are the average brain-brain coupling of individual students and student groups in the pre-class stage, the average brain-brain coupling of individual students and teacher in a natural resting state without eye contact, and the average brain-brain coupling of individual students and teacher, respectively; the third baseline parameter is the average brain-frequency following response of individual students and neural activity frequency bands in the pre-class stage in a natural resting state within a preset time. The Level I thresholds for the first, second, and third evaluation parameters are calculated as follows: and Level II threshold : Among them, when hour, This represents the first baseline parameter corresponding to the first evaluation parameter, when... hour, This represents the second baseline parameter corresponding to the second evaluation parameter, when hour, This indicates the third baseline parameter corresponding to the third evaluation parameter.

4. The classroom teaching quality assessment method based on multi-person brain-computer interface according to claim 1, characterized in that, The second assessment parameter, namely the student-teacher brain-brain coupling value, is calculated as follows: Calculate the cross-sample entropy value between the EEG signals of the currently tested student and the teacher in the same lead and frequency band; The time periods during which teachers and students engage in verbal and eye contact are defined as the time periods of interest. After obtaining the cross-sample entropy values ​​within the time periods of interest, the average values ​​are calculated to obtain the average cross-sample entropy between the current student and the teacher. After obtaining the average cross-sample entropy within a predefined region of interest, a second average is performed to obtain the second average of the cross-sample entropy.

5. The classroom teaching quality evaluation method based on multi-person brain-computer interface according to claim 2, characterized in that, The fourth assessment parameter, namely the individual student's memory efficiency, is obtained through the following method: in, This refers to the number of questions answered correctly in the post-test. This refers to the number of questions answered correctly in both the pre-test and post-test. This refers to the total number of in-class quiz questions, which is usually a constant.

6. A classroom teaching quality assessment system based on a multi-person brain-computer interface, used to execute the classroom teaching quality assessment method based on a multi-person brain-computer interface as described in any one of claims 1 to 5, characterized in that, include: The data acquisition module includes an EEG signal acquisition unit, which is used to collect EEG signals of students and teachers in a natural resting state without eye contact before class, and to collect EEG time-series signals of students and teachers during class. The voice acquisition unit is used to collect the voice signals of the teacher during the lecture; The behavior acquisition unit is used to present in-class tests and obtain behavioral data of the student group based on the test results; EEG time series signals, speech signals and behavioral data are uploaded to the real-time data stream server and then sent to the data processing module after synchronization; The data processing module performs feature extraction, pattern recognition, and fusion decision-making on EEG time-series signals, speech signals, and behavioral data during the lesson, and then outputs the teaching quality level. It also includes an assessment and feedback module, which presents the teaching quality level of all students to teachers and the teaching quality level of individual students to students.

7. A terminal, comprising a processor and a communication interface coupled to the processor, the processor being configured to run computer programs or instructions to implement the classroom teaching quality assessment method based on a multi-person brain-computer interface as described in any one of claims 1 to 5.

Images