Teaching evaluation method, program product, electronic equipment and storage medium

Abstract

The invention provides a teaching evaluation method, a program product, electronic equipment and a storage medium. The method comprises the following steps: collecting physiological data of a student in a learning activity through wearable equipment; analyzing the physiological data, and determining the learning state of the student in the learning activity; and obtaining teaching auxiliary information according to the learning state corresponding to the learning activity. According to the technical scheme, the physiological data are collected through the wearable device, the physiological data are analyzed, the learning state of the student in the learning activity is determined, then teaching evaluation is conducted, through the above objective evaluation means, the influence of subjective judgment of a teacher is reduced, and evaluation of the learning effect is more accurate and reliable. The wearable device can collect and monitor the physiological data of each student in real time, so that evaluation is more comprehensive, and omission is reduced.

Description

Technical Field

[0001] The present application relates to the field of education, and in particular, to a teaching evaluation method, a program product, an electronic device and a storage medium. Background Art

[0002] In traditional educational activities, teachers usually rely on observing students' performance in class to evaluate learning outcomes and make subsequent teaching plans based on these learning outcomes. However, the current method of evaluating learning outcomes by observation cannot observe every student in real time, and there are omissions; and this method has a large subjective influence and lacks data support or objective basis, resulting in inaccurate observation results. Summary of the invention

[0003] The purpose of the embodiments of the present application is to provide a teaching evaluation method, a program product, an electronic device and a storage medium, which are used to improve the above-mentioned defects of relying on manual observation to evaluate learning effects.

[0004] In a first aspect, an embodiment of the present application provides a teaching evaluation method, including: collecting physiological data of students in learning activities through wearable devices; analyzing the physiological data to determine the learning status of students in the learning activities; and obtaining teaching auxiliary information based on the learning status corresponding to the learning activities.

[0005] In the above implementation process, wearable devices are used to collect physiological data, analyze the physiological data, determine the learning status of students in learning activities, and then conduct teaching evaluation. Wearable devices can collect and monitor the physiological data of each student in real time, making the evaluation more comprehensive and reducing omissions. And by collecting physiological data, an objective evaluation method, the influence of teachers' subjective judgment is reduced, making the evaluation of learning effects more accurate and reliable.

[0006] Optionally, in an embodiment of the present application, the physiological data includes heart rate data; the physiological data is analyzed to determine the learning state of the student in the learning activity, including: extracting heart rate characteristics within the learning activity from the heart rate data; the heart rate characteristics include heart rate statistics, heart rate change rate, heart rate variability and / or heart rate spectrum characteristics within the course time corresponding to the learning activity; based on the heart rate characteristics and a pre-trained state analysis model, the learning state in the learning activity is obtained; the learning state includes emotional state or concentration.

[0007] In the above implementation process, a state analysis model is pre-trained that can accurately predict students' emotional state or concentration based on heart rate data, thereby providing scientific data support for teaching. After obtaining the heart rate data, the heart rate features are extracted and input into the trained state analysis model. The learning state in the learning activity is output through the model, which reduces labor costs and improves the accuracy of teaching evaluation.

[0008] Optionally, in an embodiment of the present application, the physiological data also includes derived data; the derived data includes at least one of respiratory rate, skin electrode activity, body temperature, blood pressure data, blood oxygen saturation and exercise volume; according to the heart rate characteristics and a pre-trained state analysis model, the learning state in the learning activity is obtained, including: extracting derived features in the learning activity from the derived data; the derived features include statistical values ​​of the derived data in the learning activity, the correlation between the derived data and the heart rate data, or the correlation between the derived data; the heart rate characteristics and the derived features are combined to generate physiological characteristics; the physiological characteristics are input into the pre-trained state analysis model to obtain the learning state in the learning activity.

[0009] In the above implementation process, by combining multiple physiological data, the learning status of students in different learning activities can be more comprehensively judged. The addition of derived features can improve the accuracy of the state analysis model because they provide additional information that helps to distinguish different learning states. Different physiological data may perform better in different situations. Combining heart rate features and derived features can complement each other's shortcomings and provide a more robust and comprehensive feature set, thereby analyzing a more accurate learning state.

[0010] Optionally, in an embodiment of the present application, before inputting physiological characteristics into a pre-trained state analysis model to obtain the learning state in the learning activity, the method also includes: obtaining sample heart rate data and sample derived data under different learning states; obtaining sample heart rate characteristics based on the sample heart rate data; and obtaining sample derived characteristics based on the sample derived data; the sample derived characteristics include statistical values ​​of sample derived data under different learning states, the correlation between sample derived data and sample heart rate data, or the correlation between sample derived data; combining the sample heart rate characteristics and sample derived characteristics under the same learning state, and adding a state label corresponding to the learning state to obtain a sample feature vector; the state label includes an emotion label or a concentration label; using the sample feature vector, training the machine learning model to obtain a state analysis model; the state analysis model includes an emotion analysis model or a concentration model.

[0011] In the above implementation process, combining multiple physiological data when training the state analysis model can provide more comprehensive physiological state information, thereby more accurately capturing changes in students' emotions or concentration; heart rate data and derivative data can reduce the limitations of a single data source and improve the accuracy of model predictions. By learning the correlation between multiple physiological data, the model can improve the generalization ability of the state analysis model and enable the state analysis model to be better applied to different students and conditional environments.

[0012] Optionally, in an embodiment of the present application, the teaching auxiliary information includes the student's liking for the target course; the teaching auxiliary information is obtained according to the learning status corresponding to the learning activity, including: obtaining the student's learning status corresponding to multiple learning activities of the target course, and determining the student's liking for the course according to the proportion of times the learning status is well characterized in the learning status corresponding to the multiple learning activities.

[0013] In the above implementation process, during the teaching process, the teacher can only determine the students' liking for the course through classroom performance or tests, but these all rely on external performance and are prone to overlooking some unexpressed advantages and preferences. The embodiment of the present application determines the students' liking for the course based on the proportion of times the learning state is well characterized in the learning state corresponding to multiple learning activities, reduces the deviation of subjective judgment, and provides a more objective way to evaluate liking. And it can identify students' interest in the target course before it is shown in the early stage or when the students themselves are not aware of it, and tap into students' interests more accurately and efficiently.

[0014] Optionally, in an embodiment of the present application, the teaching auxiliary information includes the students' learning effect on the course corresponding to the learning activity; the teaching auxiliary information is obtained according to the learning status corresponding to the learning activity, including: determining the students' learning effect on the course corresponding to the learning activity according to the students' learning status corresponding to the learning activity, and providing personalized homework guidance based on the learning effect; wherein, the learning status and learning effect are positively correlated.

[0015] In the above implementation process, a good learning state and learning effect are inseparable. Through qualitative or quantitative analysis of the learning state, the learning effect of the learning activities can be judged, and personalized homework guidance can be provided, and personalized learning resources and activities can be provided to make it more in line with students' interests and needs. When the course content matches the students' interests, the students' participation and interactivity will also increase, which is conducive to the students' all-round development.

[0016] Optionally, in an embodiment of the present application, the teaching auxiliary information includes a visual chart; the teaching auxiliary information is obtained according to the learning status corresponding to the learning activity, including: drawing a visual chart based on the learning status of students in different learning activities obtained in real time; the axes of the visual chart are respectively the time and learning status corresponding to the learning activity; the visual chart is used as a reference object for adjusting the teaching rhythm.

[0017] In the above implementation process, a visualization chart is drawn based on the real-time learning status of students in different learning activities. The visualization chart can intuitively display the learning status of students at different times, visualize the learning status, and enable teachers to quickly and accurately grasp the changes in students' learning status. The real-time updated chart provides teachers with instant feedback, allowing them to adjust teaching strategies in a timely manner to meet students' needs.

[0018] Optionally, in an embodiment of the present application, the teaching-auxiliary information includes content that students need targeted tutoring; the content that students need targeted tutoring is the content that students need targeted tutoring; the teaching-auxiliary information is obtained according to the learning status corresponding to the learning activity, including: determining the target time when the student's learning status is not good according to the student's corresponding learning status in different learning activities, and determining the content that students need targeted tutoring from the teaching audio / video according to the target time; the teaching audio / video is obtained by recording the teacher's teaching process to obtain the content that students need targeted tutoring.

[0019] In the above implementation process, according to the corresponding learning status of students in different learning activities, the target time when the student's learning status is not good is determined, and the content that the student needs to learn in a targeted manner is determined from the teaching audio / video according to the target time. In this way, teachers can more accurately identify students' difficulties in specific knowledge points and provide timely and targeted tutoring, thereby improving students' learning effects. At the same time, this method also helps to better understand students' learning process, optimize teaching strategies, and improve teaching quality.

[0020] Optionally, in an embodiment of the present application, the teaching auxiliary information includes information representing the learning status of the class; the teaching auxiliary information is obtained according to the learning status corresponding to the learning activity, including: generating information representing the learning status of the class according to the learning status of multiple students participating in the course, and sending the information representing the learning status of the class to the teacher terminal.

[0021] In the above implementation process, the information representing the class learning status is sent to the teacher's terminal. The teacher can check the learning status of the students in the class and understand the real-time status changes of the students in the course. By sending the information representing the class learning status to the teacher's terminal, a good feedback loop is established. The teacher can adjust the teaching rhythm according to the students' learning status. The students' performance in turn affects the adjustment of teaching content and methods, which helps teachers optimize teaching strategies and improve teaching quality.

[0022] Optionally, in an embodiment of the present application, after collecting physiological data of students in learning activities through wearable devices, the method also includes: obtaining physiological indicator change data of students participating in the same course based on the physiological data of students participating in the same course; the physiological indicator change data represents the physiological data change characteristics of students in the learning activity; based on the physiological indicator change data of students participating in the same course, determining target students from students participating in the same course; the target students are students whose learning status characteristics are abnormal.

[0023] In the above implementation process, target students (i.e. students with abnormal learning status) are identified based on the physiological indicator change data of students participating in the same course, and teachers can provide personalized guidance and support to these students; and by analyzing the changes in physiological indicators of the group, the impact of common external stimuli on students can be identified. This group behavior analysis helps to understand classroom dynamics, help teachers adjust teaching strategies, and use group reactions to design more attractive teaching activities.

[0024] Optionally, in an embodiment of the present application, after analyzing the physiological data to determine the student's learning status in the learning activity, the method also includes: when it is determined through the physiological data that the student's emotional state represents fear, judging whether the student is in a dangerous state based on at least one of the acceleration data, gyroscope data, audio data and video data collected by the wearable device; if so, generating an alarm message and sending the alarm message to a preset terminal in real time; the alarm message includes location information.

[0025] In the above implementation process, the emotional state is first judged, and then the various data types collected by wearable devices are combined with analysis technology to achieve real-time monitoring and early warning of students' dangerous conditions, which helps schools take timely measures to protect students' safety and reduce potential safety risks.

[0026] Optionally, in an embodiment of the present application, the physiological data includes at least one of heart rate data, respiratory rate, skin electrode activity, body temperature, blood pressure data, blood oxygen saturation and exercise volume; collecting the physiological data of students in learning activities through wearable devices includes: collecting the physiological data of students in learning activities through wearable devices according to the course type corresponding to the learning activity; the course type includes at least one of cultural course type, physical education course type, art course type and practical course type.

[0027] In the above implementation process, the collection module configuration of the wearable device is dynamically adjusted according to the course type to realize the intelligent collection and management of students' physiological data, which can not only improve the pertinence and accuracy of physiological data collection, but also effectively save power and extend the use time of wearable devices.

[0028] Optionally, in an embodiment of the present application, physiological data is analyzed to determine the learning status of students in learning activities, including: when the course type is a cultural course type, if the amount of exercise in the physiological data exceeds a preset threshold, the learning status is determined to be poor; or, if the physiological data represents the student's static state, and the static state exceeds a preset duration, the learning status is determined to be poor.

[0029] In the above implementation process, the learning status of students in the process of learning cultural courses is evaluated by monitoring their physiological data, and the physiological data of students, such as heart rate, activity level, etc., are collected and analyzed to judge whether the learning status of students is in a bad state. The learning status of students in the process of learning cultural courses and physical education courses can be understood more accurately, so that effective intervention measures can be taken in time to improve students' learning efficiency and effect.

[0030] Optionally, in an embodiment of the present application, the learning status includes emotional state or concentration. After analyzing the physiological data and determining the student's learning status in the learning activity, the method also includes: generating status information based on the student's learning status in the learning activity; and sending the status information to the student's parent terminal.

[0031] In the above implementation process, parents can understand their children's performance in school in real time and enhance communication and cooperation with teachers. By establishing a more open, transparent and interactive educational environment, parents can become active participants in their children's education and jointly promote their children's learning and growth.

[0032] Optionally, in an embodiment of the present application, generating status information according to the learning status of the student in the learning activity includes: generating status information according to the course time and learning status corresponding to the learning activity; the status information includes learning status corresponding to different times and / or learning status corresponding to different courses. Wherein, time refers to a time period or a time point.

[0033] In the above implementation process, parents can view students' learning status at different times and / or different courses in real time, pay attention to changes in students' learning status in a timely manner, provide help and support, and understand students' status at each time or in each class more clearly and conveniently with finer granularity, which helps to evaluate changing trends in students' learning habits, stress levels and health conditions.

[0034] On the second aspect, the embodiment of the present application also provides a teaching evaluation device, including: a collection module, used to collect physiological data of students in learning activities through wearable devices; an analysis module, used to analyze the physiological data and determine the learning status of students in learning activities; an evaluation module, used to obtain teaching auxiliary information based on the learning status corresponding to the learning activities.

[0035] In a third aspect, an embodiment of the present application further provides a computer program product, including computer program instructions, which, when executed by a processor, execute the method provided by the first aspect or any one of the implementations of the first aspect.

[0036] In a fourth aspect, an embodiment of the present application further provides an electronic device, comprising: a processor and a memory, the memory storing computer program instructions, and the computer program instructions, when executed by the processor, execute the method provided by the first aspect or any one of the implementations of the first aspect.

[0037] In a fifth aspect, an embodiment of the present application further provides a computer-readable storage medium, on which computer program instructions are stored. When the computer program instructions are executed by a processor, the method provided by the first aspect or any one of the implementations of the first aspect is executed.

[0038] By using a teaching evaluation method, program product, electronic device and storage medium provided in the embodiment of the present application, physiological data is collected through a wearable device, the physiological data is analyzed, the learning status of students in learning activities is determined, and then teaching evaluation is performed. Through the above objective evaluation means, the influence of the teacher's subjective judgment is reduced, making the evaluation of learning effects more accurate and reliable. Wearable devices can collect and monitor the physiological data of each student in real time, making the evaluation more comprehensive and reducing omissions. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the embodiments of the present application will be briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present application and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0040] Figure 1 A flow chart of a teaching evaluation method provided in an embodiment of the present application;

[0041] Figure 2 A schematic diagram of the structure of a teaching evaluation device provided in an embodiment of the present application;

[0042] Figure 3 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0043] The following embodiments of the technical solution of the present application are described in detail in conjunction with the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solution of the present application, and are therefore only used as examples, and cannot be used to limit the scope of protection of the present application.

[0044] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by technicians in the technical field to which this application belongs; the terms used herein are only for the purpose of describing specific embodiments and are not intended to limit this application.

[0045] In the description of the embodiments of the present application, the technical terms "first", "second", etc. are only used to distinguish different objects, and cannot be understood as indicating or implying relative importance or implicitly indicating the number, specific order or primary and secondary relationship of the indicated technical features. In the description of the embodiments of the present application, the meaning of "multiple" is two or more, unless otherwise clearly and specifically defined.

[0046] In traditional educational activities, teachers usually rely on observing students' performance in class to evaluate learning outcomes and make subsequent teaching plans based on these learning outcomes, such as recommending courses for individuals, exploring their interests, and assigning homework, etc. This approach has many limitations.

[0047] First, in a teaching environment with a large number of students, it is difficult for teachers to pay full attention to the status of each student. Due to time and attention constraints, teachers may not be able to fully evaluate the learning effect of each student. Although teachers will try their best to observe the reactions and participation of each student, in a classroom environment with limited time, it is inevitable that some students' performance will be overlooked. This situation may lead to an incomplete evaluation of the learning effect of individual students, and may even miss some potential problems.

[0048] Secondly, teachers’ subjective evaluation is easily affected by their personal emotions, experience and perspective. Each teacher’s observation ability and judgment criteria may be different, resulting in certain deviations in the evaluation results. For example, some students may not be easily detected for their learning difficulties due to their introverted personality, or active students may mask their actual lack of knowledge. This subjectivity makes the evaluation of learning effects often lack objective basis and cannot truly reflect the students’ learning status.

[0049] In summary, although the traditional observational evaluation method provides feedback to teachers and parents to a certain extent, it leads to inaccurate and incomplete evaluation of learning effects due to factors such as class size, subjective bias, and time constraints. Therefore, it is necessary to introduce a more scientific and systematic evaluation method to ensure that the learning effect of each student is accurately reflected, so as to achieve personalized educational support and effective learning guidance.

[0050] This application example provides a teaching evaluation method, which collects students' physiological data during learning activities through wearable devices; analyzes the physiological data to determine the students' learning status during learning activities, and then conducts teaching evaluation. Through the above objective evaluation means, the influence of teachers' subjective judgment is reduced, making the evaluation of learning effects more accurate and reliable. Wearable devices can collect and monitor the physiological data of each student in real time, making the evaluation more comprehensive and reducing omissions.

[0051] See also Figure 1 A flow chart of a teaching evaluation method provided by an embodiment of the present application is shown. The teaching evaluation method provided by an embodiment of the present application can be applied to an electronic device, which may include a physical device such as a wearable device, a server, a PC, a tablet computer, or a smart phone, or a virtual device such as a virtual machine or a container. The electronic device may be a single device, or a combination of multiple devices or a cluster of a large number of devices. The teaching evaluation method may include:

[0052] Step S110: Collecting physiological data of students during learning activities through wearable devices.

[0053] Step S120: Analyze the physiological data to determine the learning status of the student in the learning activity.

[0054] Step S130: Obtain teaching auxiliary information according to the learning status corresponding to the learning activity.

[0055] In step S110: Wearable devices, also known as wearable electronic devices or wearable computers, refer to smart devices that can be worn directly on the body (such as the head, wrist, chest, etc.). Wearable devices can be devices such as smart watches, smart bracelets or smart vests.

[0056] Learning activities can refer to the process of students attending classes, doing sports (including fitness or physical education classes) or self-learning. Physiological data refers to various data generated by the human body during physiological activities, such as heart rate data, respiratory rate, skin electrode activity, body temperature, blood pressure data, blood oxygen saturation and / or exercise volume.

[0057] The ways in which wearable devices can collect physiological data include: (1) Wearable devices can be equipped with physiological sensors, which can monitor students' vital signs in real time, including heart rate data, respiratory rate, body temperature, blood pressure data and blood oxygen saturation, etc.; (2) Wearable devices can also be equipped with motion sensors, which can detect students' exercise volume, such as number of steps, acceleration, angle, step frequency, calorie consumption, exercise time, exercise type (such as aerobic exercise or anaerobic exercise), exercise density (referring to the proportion of actual exercise time per unit time), etc. In addition to quantitative indicators, the quality of exercise (such as accuracy, continuity, and coordination of movements) is also an important aspect of evaluating exercise volume. These data are helpful in analyzing students' activity volume and exercise intensity; (3) Wearable devices can also be equipped with bioinformatics sensors, which can detect students' skin electrical activity, brain wave information, heart rate variability, etc.

[0058] It should be noted that the process implemented in the embodiment of the present application is to collect students' physiological data during learning activities in strict compliance with relevant laws, regulations and ethical norms and with the consent of the students themselves or their guardians.

[0059] In step S120: the physiological data is analyzed by using a trained state analysis model, a statistical method or a preset rule. The learning state may include the emotional state or concentration in the learning activity.

[0060] Among them, the state analysis model can identify characteristic patterns related to the learning state by learning a large amount of physiological data. For example, in emotion recognition and concentration judgment, the neural network can analyze physiological signals such as heart rate, skin conductance, and brain waves to infer the student's emotional state and concentration. The process of using the trained state analysis model for analysis and the generation process of the state analysis model will be introduced in more detail later.

[0061] Statistical analysis methods include descriptive statistics, variance analysis, correlation analysis, and regression analysis. Statistical analysis methods can quantitatively describe physiological data, compare the differences in physiological indicators under different learning states, explore the relationship between physiological data and learning states, and predict the changing trend of physiological data under different learning states. Taking emotional state as an example, changes in heart rate can reflect the student's tension or excitement. In positive emotions, the heart rate may rise slightly; while in anxiety or fear, the heart rate may increase significantly. Therefore, the student's emotional state and emotional changes can be determined by calculating the change in heart rate data and the preset change threshold, which can be used as the learning state. Heart rate variability (HRV) refers to the change in the heartbeat interval, which can reflect the activity of the autonomic nervous system. High HRV usually indicates a state of relaxation and recovery, while low HRV may reflect a state of tension or stress. High HRV may indicate that the student remains calm and focused when studying, while low HRV may indicate stress, fatigue, or distraction.

[0062] For example, changes in body temperature may reflect changes in emotions and physiological states. Excitement, tension, or increased stress can cause body temperature to rise. By monitoring body temperature fluctuations, it is possible to determine whether students are in an excited or nervous state, and combine other physiological data to assess their learning status. Blood oxygen saturation refers to the oxygen content in the blood, which is usually used to assess students' physical condition and fatigue level. Lower blood oxygen levels may reflect fatigue or inefficient learning states, thereby determining concentration, especially during long periods of learning, and can also assist in assessing students' attention span. The above physiological data can be used to determine students' emotional state or concentration during learning activities through statistical methods.

[0063] The following is an illustrative introduction to several methods of analyzing learning status using preset rules. The preset rules are based on physiological data and can determine the emotional state or concentration corresponding to the physiological data. The rules can be in the form of formulas, text, tables, or other data forms that represent mapping relationships. For example, the heart rate and breathing rate can be combined to analyze the emotional state. The synchronous increase in heart rate and breathing rate is usually associated with positive emotional states, such as pleasure and excitement. Rule: When the heart rate reaches a first threshold (such as 80-100 beats per minute, representing a state of more emotional excitement or excitement), and at the same time the breathing rate exceeds a second threshold (such as 16-20 beats per minute, indicating shortness of breath), it may represent an emotion of pleasure, excitement, or excitement.

[0064] For another example, stable physiological indicators are usually related to the individual's state of concentration and relaxation, which helps maintain clear thinking and efficient cognitive activities. The rule can be: when the heart rate and respiratory rate remain relatively stable (heart rate between 60-80 beats per minute, respiratory rate between 12-16 beats per minute), and the fluctuation range is small (such as heart rate fluctuation does not exceed 10 times / minute, respiratory rate fluctuation does not exceed 2 times / minute), it may indicate that the individual is in a state of high concentration.

[0065] It can also be that blood oxygen saturation remains within the normal range (such as 95%-100%) and has no obvious fluctuations (fluctuations are less than the threshold), which may indicate that the individual is breathing steadily and has sufficient oxygen supply when concentrating. Blood oxygen saturation is an important indicator of an individual's respiratory function and oxygen supply status. Stable blood oxygen saturation helps maintain efficient cognitive activities.

[0066] The following are some examples of rules that require multiple physiological data to participate. These rules take into account multiple physiological indicators such as heart rate data, breathing rate, skin electrodermal activity, body temperature, blood pressure data, blood oxygen saturation, and exercise volume to more accurately assess emotional state, concentration, or other physiological and psychological states:

[0067] When the heart rate is at a higher level than the normal range (such as 80-100 times per minute), the breathing rate is slightly faster but stable (such as 16-20 times per minute), the skin electrical activity rises slightly but does not exceed the upper limit of the normal range, the blood oxygen saturation remains above 95%, and the amount of exercise is reduced, it may indicate that the individual is in a state of pleasure. When the heart rate is significantly accelerated (such as more than 100 times per minute), the breathing rate is rapid (such as more than 24 times per minute), the skin electrical activity rises significantly (exceeding a certain percentage of the normal range), the blood pressure rises slightly, the blood oxygen saturation may drop slightly, and the amount of exercise increases (such as frequent walking and more limb movements), it may indicate that the individual is in an anxious state.

[0068] When the heart rate remains relatively stable (such as 60-80 times per minute), the respiratory rate is stable (such as 12-16 times per minute), the skin electrical activity is in the normal range, the blood oxygen saturation remains above 95%, the body temperature drops slightly (possibly due to the slowing of body metabolism caused by concentration), the blood pressure is stable, and the amount of exercise is significantly reduced (such as reduced limb movements and reduced head rotation), it may indicate that the individual is in a state of high concentration. Furthermore, when the heart rate fluctuates greatly (such as exceeding the fluctuation range of 10 times per minute), the respiratory rate is unstable (such as frequent deep breathing or shallow breathing), the skin electrical activity fluctuates frequently, the blood oxygen saturation drops slightly, the body temperature is normal or slightly high, the blood pressure may rise, and the amount of exercise increases (such as frequent walking and increased limb movements), it may indicate that the individual is in a state of low concentration.

[0069] It should be noted that the use of rules to analyze learning status should take individual differences into consideration. Different individuals have different physiological responses, so the above rules may need to be adjusted and optimized according to specific individuals.

[0070] In step S130: After determining the learning status, teaching auxiliary information can be obtained according to the learning status. Teaching auxiliary information refers to various data, analysis results and visual display materials used to help teachers better understand students' learning situations, optimize teaching methods and improve teaching effectiveness during the teaching process. Teaching auxiliary information provides teachers with comprehensive, real-time and personalized teaching support. Teaching auxiliary information includes, for example, students' learning effects on the course corresponding to this learning activity; or students' liking or interest in courses of this type; teaching auxiliary information can also be visual charts reflecting students' learning status in different time periods, etc.

[0071] For example, if the learning status is the emotional state of the students, if the students' emotional state is good, the students may be more receptive to the learning activities, which means that the learning activities are more effective and the students' learning effects are more significant. On the contrary, if the students' emotional state is not good, the students may not have a good learning effect on the learning activities (courses), and they may need to reflect on and improve the learning activities. Similarly, if the learning status is concentration, the higher the concentration, the better the students' learning effects.

[0072] In the implementation process of the above embodiment: physiological data is collected through wearable devices, and the physiological data is analyzed to determine the learning status of students in learning activities, and then teaching evaluation is performed. The wearable device can collect and monitor the physiological data of each student in real time, making the evaluation more comprehensive and reducing omissions. And by collecting physiological data, an objective evaluation method, the influence of teachers' subjective judgment is reduced, making the evaluation of learning effects more accurate and reliable.

[0073] Optionally, in an embodiment of the present application, the physiological data includes heart rate data. For example, if the learning activity is a class, the real-time heart rate of the student during the class is collected through a wearable device. Analyzing the physiological data to determine the learning status of the student in the learning activity includes the following steps:

[0074] (1) Extracting heart rate features in learning activities from heart rate data; heart rate features include heart rate statistics, heart rate change rate, heart rate variability and / or heart rate spectrum features during the course time corresponding to the learning activity. Before extracting heart rate features, the collected heart rate data can be cleaned, including removing outliers, missing values, etc., to ensure the accuracy and completeness of the data.

[0075] Heart rate statistics can include average heart rate, maximum heart rate or minimum heart rate. The average heart rate value reflects the student's overall heart rate level during the entire course; the maximum heart rate reflects the student's maximum load during the course; the minimum heart rate may occur during activity intervals or rest periods.

[0076] The heart rate change rate includes the heart rate rise rate or the heart rate fall rate. For example, the heart rate rise rate is obtained by calculating the increase in heart rate from one moment to another and dividing it by the time interval. The heart rate fall rate is obtained by calculating the decrease in heart rate from a peak moment to a subsequent moment and dividing it by the time interval.

[0077] Heart rate variability includes heart rate standard deviation or heart rate variability index. For example, the standard deviation of heart rate data is calculated, which reflects the degree of fluctuation of heart rate at different time points. A larger standard deviation indicates higher heart rate variability, which may mean that the student's heart is more adaptable to activities or has larger physiological fluctuations. The heart rate variability index is calculated through more complex algorithms (such as time series analysis, frequency domain analysis, etc.) to more accurately quantify the volatility and stability of heart rate.

[0078] Heart rate spectrum features refer to spectrum analysis of heart rate data, which decomposes the heart rate signal into a superposition of different frequency components. By analyzing the distribution and intensity of these frequency components, the heart rate spectrum features can be extracted.

[0079] (2) According to the heart rate characteristics and the pre-trained state analysis model, the learning state in the learning activity is obtained; the learning state includes the emotional state or the degree of concentration. The state analysis model can be obtained by training based on the sample heart rate characteristics, or by training based on the sample heart rate data and the sample derived data. The process of obtaining the state analysis model by training based on the sample heart rate characteristics is described below.

[0080] Use smart bracelets or other wearable devices to collect a large number of students' heart rate data in different learning activities as sample heart rate data, or directly obtain heart rate data from historical data as sample heart rate data. Then clean the sample heart rate data to remove outliers and noise. Use the processed sample heart rate data for feature engineering to extract sample heart rate statistics, sample heart rate change rate, sample heart rate variability and / or sample heart rate spectrum features. It should be noted here that the way to extract sample heart rate features when building a state analysis model should be the same as the way to extract heart rate features in learning activities from heart rate data during subsequent state analysis. For example, when building a state analysis model, the way to extract sample heart rate features is to extract the sample heart rate change rate. Then, the heart rate change rate should also be extracted as a heart rate feature during subsequent state analysis.

[0081] Determine the state label of the emotional state or concentration, which can be determined through teacher observation, student self-report or other tests. For example, the emotional label can be positive or negative, or it can be more detailed, including happy, joyful, excited, excited, nervous, depressed or angry, etc., of course, it is a numerical value within a preset range. The larger the numerical value, the more positive the emotion, and the lower the numerical value, the more negative the emotion. The concentration label can also be a number. The larger the numerical value, the higher the concentration, and the lower the numerical value, the lower the concentration. Assign a corresponding state label (emotion label or concentration label) to each sample heart rate feature to form a training set. Select a suitable machine learning algorithm, such as support vector machine (SVM), random forest, neural network, etc. For deep learning methods, a multi-layer perceptron (MLP) or convolutional neural network (CNN) can be constructed. Use the labeled training set to train the model, adjust the model parameters to optimize the performance, and obtain the state analysis model. You can also use cross-validation and other techniques to evaluate the generalization ability of the state analysis model until it meets the requirements.

[0082] The heart rate characteristics are input into the above-mentioned trained state analysis model, and the learning state in the learning activity is output through the model.

[0083] In the implementation process of the above embodiment: a state analysis model is pre-trained that can accurately predict the emotional state or concentration of students based on heart rate data, thereby providing scientific data support for teaching. After obtaining the heart rate data, the heart rate features are extracted, and the heart rate features are input into the trained state analysis model. The learning state in the learning activity is output through the model, which reduces labor costs and improves the accuracy of teaching evaluation.

[0084] Optionally, in the embodiment of the present application, the physiological data also includes derived data; the derived data includes at least one of respiratory rate, skin electrical activity, body temperature, blood pressure data, blood oxygen saturation and exercise volume; skin electrical activity is such as skin conductance level. According to the heart rate characteristics and the pre-trained state analysis model, the learning state in the learning activity is obtained, including:

[0085] Extract derived features in learning activities from derived data; derived features include derived data statistics in learning activities, correlations between derived data and heart rate data, or correlations between derived data. Derived data statistics include the average, maximum, or minimum values ​​of derived data. Taking respiratory rate as an example, the average respiratory rate of students in the course time corresponding to the learning activity can be counted as a derived feature. Derived data statistics can reflect the changes in students' learning status over a period of time, which helps to dynamically monitor students' learning progress and efficiency. By analyzing the correlation between derived data and heart rate data, the interaction between different physiological indicators can be revealed, providing more perspectives for understanding the changes in status during the learning process.

[0086] The heart rate features and derived features are combined to generate physiological features. The combination can be done by splicing. If the heart rate features and derived features are both numerical values, the physiological features can also be generated by adding, multiplying, or taking a ratio. It can be understood that the heart rate features and derived features are both features in the same learning activity. For example, the heart rate data and derived data collected during the same course are subjected to feature extraction, and the heart rate features and derived features during the same course are combined to generate physiological features. The physiological features are then input into a pre-trained state analysis model to obtain the learning state in the learning activity.

[0087] In the implementation process of the above embodiment: by combining multiple physiological data, the learning status of students in different learning activities can be more comprehensively judged. The addition of derived features can improve the accuracy of the state analysis model, because the derived features provide information other than heart rate data, which helps to distinguish different learning states. Different physiological data may perform better in different situations. Combining heart rate features and derived features can complement each other's shortcomings and provide a more robust and comprehensive feature set, thereby analyzing a more accurate learning state.

[0088] Optionally, in an embodiment of the present application, before inputting physiological characteristics into a pre-trained state analysis model to obtain the learning state in the learning activity, the method also includes the step of training the state analysis model. The process of training the state analysis model is described below.

[0089] Sample heart rate data and sample derived data under different learning states are obtained. For example, if the state analysis model is an emotion analysis model, sample heart rate data and sample derived data under positive emotions, and sample heart rate data and sample derived data under negative emotions can be obtained by questionnaire survey, observation or testing. If the state analysis model is a concentration model, sample heart rate data and sample derived data corresponding to different concentrations (such as high concentration, medium concentration and low concentration; or different concentration values) can be obtained by questionnaire survey, observation or testing.

[0090] The sample heart rate feature is obtained according to the sample heart rate data, and the sample derived feature is obtained according to the sample derived data; the sample derived feature includes the statistical value of the sample derived data under different learning states, the correlation between the sample derived data and the sample heart rate data, or the correlation between the sample derived data. Please refer to the above for the method of extracting the sample heart rate feature. Taking the sample derived feature as the correlation between the sample derived data and the sample heart rate data as an example, if the sample derived data is skin electrode activity, the sample skin electrode activity data and the sample heart rate data are collected respectively under different states of the same subject, and the appropriate correlation calculation method is selected according to the nature and characteristics of the data. Commonly used methods include Pearson correlation coefficient, Spearman rank correlation coefficient, etc. The selected method can also be used to calculate the correlation coefficient between the heart rate and the skin conductance level as the correlation relationship, which can quantify the linear relationship or rank correlation between the two. The correlation algorithm result of the skin electrode activity data and the sample heart rate data or the correlation coefficient between the two can be used as a sample derived feature.

[0091] The sample heart rate features and sample derived features under the same learning state are combined, and the state label corresponding to the learning state is added to obtain the sample feature vector; the state label includes an emotion label or a concentration label. It should be noted that the combination of sample heart rate features and sample derived features when constructing the state analysis model should be consistent with the combination of heart rate features and derived features when analyzing the learning state.

[0092] The sample feature vector is used to train the machine learning model to obtain a state analysis model; the state analysis model includes a sentiment analysis model or a concentration model. The machine learning model or the neural network model can be used to train and test the sample feature vector containing the derived features to obtain the state analysis model.

[0093] In the implementation process of the above embodiment: when training the state analysis model, combining multiple physiological data can provide more comprehensive physiological state information, thereby more accurately capturing the changes in students' emotions or concentration; heart rate data and derivative data can reduce the limitations of a single data source and improve the accuracy of model prediction. By learning the correlation between multiple physiological data, the model can improve the generalization ability of the state analysis model, so that the state analysis model can be better applied to different students and conditional environments.

[0094] Optionally, in an embodiment of the present application, the teaching-auxiliary information includes the student's liking for the target course, which can reflect the student's interest in a certain subject. For example, a relatively stable heart rate and high concentration in a target cultural course may indicate that the student is more interested in or is more proficient in the course; and in a target experience course, if the heart rate reaches a specific range and is accompanied by a sense of excitement, it can be inferred that the student has a high interest in sports.

[0095] According to the learning status corresponding to the learning activities, teaching auxiliary information is obtained, including: obtaining the learning status corresponding to multiple learning activities of students for the target course, and determining the students' liking for the course according to the proportion of times the learning status is well represented in the learning status corresponding to the multiple learning activities.

[0096] For example, the learning status corresponding to each course in a preset time period (such as a month or a week, etc.) can be obtained, the target course is a math class, and the learning status of the math class in the preset time period is counted, and the total number of math classes in the preset time period is 10 times, among which the number of times the learning status of the math book is well represented is 8 times, and the number of times the learning status is well represented is 8 / 10, which exceeds the preset proportion (for example, it can be set to 1 / 2), then it is determined that the student likes the course highly. Among them, the target course can be any course, and the good learning status can be represented by a positive emotional state, a medium or above concentration, or an emotional score greater than a preset score.

[0097] In the implementation process of the above embodiment: During the teaching process, teachers can only determine students' liking for the course through classroom performance or tests, but these all rely on external performance and are prone to overlooking some unexpressed advantages and preferences. The embodiment of the present application determines students' liking for the course based on the proportion of times the learning state is well characterized in the learning state corresponding to multiple learning activities, reduces the bias of subjective judgment, and provides a more objective way to evaluate liking. And it can identify students' interest in the target course before it is shown in the early stage or when the students themselves are not aware of it, and tap into students' interests more accurately and efficiently.

[0098] Optionally, in an embodiment of the present application, the teaching auxiliary information includes the student's learning effect on the course corresponding to the learning activity, and the learning effect refers to the learning effect of the current course.

[0099] According to the learning status corresponding to the learning activity, teaching auxiliary information is obtained, including: according to the learning status of the students in the learning activity, determining the learning effect of the students on the course corresponding to the learning activity, and providing personalized homework guidance according to the learning effect; wherein the learning status and the learning effect are positively correlated.

[0100] For example, the current learning activity is a math class. First, the student's learning status in the math class is obtained. The learning status here can be the emotional state or concentration in the math class. If the student's learning status in the math class is well represented, for example, the emotional state is relatively pleasant, the concentration is above average, or the emotional score is greater than the preset score, then it can be judged that the student has learned well in this math class; if the student's learning status in the math class is not well represented, for example, the emotional state is unpleasant, the concentration is below average, or the emotional score is less than the preset score, then it may mean that the student's absorption ability for this math class is weak, and it can be judged that the student's learning effect in this math class is not good.

[0101] Personalized homework guidance based on learning outcomes For example, teachers can assign homework of different difficulty and emphasis to different students, or arrange targeted extracurricular tutoring to help them make up for their weak points in knowledge. Students can also be divided into different study groups based on their learning outcomes. For example, the study group with good learning outcomes and high mastery can explore more difficult content in depth, while the study group with poor learning outcomes and low mastery can review or practice basic knowledge.

[0102] For another example, if the learning activity is a basketball class, first obtain the student's learning status in the basketball class. The learning status here can be the emotional state or concentration during the basketball class. If the emotional state represents happiness or high concentration, the student may be interested in the basketball class or enjoy the fun brought by basketball. For such students, more basketball sports can be carried out to help students develop their potential and keep students physically and mentally healthy in basketball. If the emotional state represents unpleasantness or low concentration, the student may not be interested in the basketball class or has not participated in basketball activities. For such students, other sports can be recommended to them, and the reasons for the student's unpleasant emotions or low concentration can be paid attention to in time, and emotional comfort or help can be provided to the student to avoid more serious consequences if this state continues.

[0103] As an implementation method, if the number of times a student's learning effect is poor in continuous learning activities exceeds a preset number, or the proportion of such poor performance in the learning activities is too large, a visualization of the learning effect can be sent to parents so that they can understand the student's learning dynamics.

[0104] In the implementation process of the above embodiment: a good learning state and learning effect are inseparable. By conducting qualitative or quantitative analysis of the learning state, the learning effect of the learning activity can be judged, and personalized homework guidance can be provided, and personalized learning resources and activities can be provided to make it more in line with students' interests and needs. When the course content matches the students' interests, the students' participation and interactivity will also increase, which is conducive to the students' all-round development.

[0105] Optionally, in an embodiment of the present application, the teaching auxiliary information includes a visual chart; and the teaching auxiliary information is obtained according to the learning state corresponding to the learning activity, including:

[0106] Draw a visualization chart based on the real-time learning status of students in different learning activities; the axes of the visualization chart are the time and learning status corresponding to the learning activities; the visualization chart is used as a reference for adjusting the teaching rhythm.

[0107] Exemplarily, a visualization chart is drawn according to the time of the learning activity and the corresponding learning status in the learning activity. The visualization chart can be a two-dimensional chart, including a line chart, a bar chart or a scatter chart, etc. The horizontal axis represents the time corresponding to the learning activity (for example, the beginning to the end of a class, or the course time of a day, or the course time of a week), and the vertical axis represents the learning status (which can be a concentration score or an emotional label of an emotional state). Each data point on the chart represents the learning status of a student at a specific time point (such as a class, or a certain hourly time point). Different colors or icons can be used for different learning statuses to intuitively distinguish the changes in students' status in different time periods. Trend lines or areas can also be added to show the overall trend and pattern of students' learning status. Teachers can view the chart in real time to understand the changes in students' learning status in the course and make timely teaching adjustments. The adjustment plan will be introduced later.

[0108] It should be noted that the content in the visualization chart can be updated in real time according to the received learning status.

[0109] In the implementation process of the above embodiment: according to the learning status of students in different learning activities obtained in real time, a visualization chart is drawn, which can intuitively show the learning status of students at different time points, visualize the learning status, and enable teachers to quickly and accurately grasp the changes in students' learning status. The real-time updated chart provides instant feedback to teachers, allowing them to adjust teaching strategies in time to meet students' needs.

[0110] Optionally, in an embodiment of the present application, the teaching auxiliary information includes content that the student needs targeted tutoring; the content that the student needs targeted tutoring is the content that the student needs targeted tutoring; and obtaining the teaching auxiliary information according to the learning state corresponding to the learning activity includes:

[0111] According to the students' corresponding learning status in different learning activities, the target time when the students' learning status is poor is determined, and the content that the students need targeted tutoring is determined from the teaching audio / video according to the target time; the teaching audio / video is obtained by recording the teacher's teaching process.

[0112] Record the time of different learning activities and the corresponding learning status of students in the learning activities, for example, the time of different learning activities, such as the start time to the end time of a class, or the time of all courses in a day, or the time of courses in a week. The learning status can be a score of concentration or an emotional label of emotional state. Poor learning status may include emotional state representation of unpleasantness, concentration being below average, concentration score less than a preset score, or emotional score less than a preset score.

[0113] Determine the time when the learning status is not good as the target time. The target time can be an exact time point (for example, 10:10 on September 15, 2020, when the concentration is below average) or a time period (for example, 10:10 on September 15, 2020 - 11:10 on September 15, 2020, when the concentration is below average).

[0114] The teaching audio / video is obtained by recording the teacher's teaching process. There is a specific timestamp in the teaching audio / video. Following the above embodiment, after determining the target time (for example, 10:10 on September 15, 2020 - 11:10 on September 15, 2020), the audio / video clip corresponding to the above time period is found based on the timestamp in the teaching audio / video, and the video clip is used as the content that students need targeted tutoring for students to conduct targeted tutoring. Students can learn this part of the content again based on the video clip to consolidate the knowledge that they have not mastered firmly.

[0115] For example, after finding the audio / video clips, we can also extract knowledge points based on the audio clips and use them as the content that students need to learn in a targeted manner. For example, we can perform keyword recognition based on the video audio and determine the content that students need to learn in a targeted manner based on the recognized keywords. For example, if we recognize "In this class, we will learn XX theorem", we will use XX theorem as the content that students need to learn in a targeted manner; or if we recognize "Turn to page XX in the Chinese textbook", we will use page XX and nearby pages as the content that students need to learn in a targeted manner, etc. If the teaching audio / video is an image or video clip, we can also use OCR to recognize content such as titles as the content that students need to learn in a targeted manner.

[0116] As another implementation, the content that the student needs to learn specifically includes the video clip and / or related content determined based on the video clip, wherein the related content includes homework for the knowledge point corresponding to the video clip, or remedial content that can improve the knowledge point, etc.

[0117] As an implementation method, the target time when the learning status is not good can also be determined based on the visualization chart obtained above. The horizontal axis of the visualization chart represents the time corresponding to the learning activity, and the vertical axis represents the learning status. According to the learning status displayed in the vertical axis, the target time when the learning status is not good can be determined based on the horizontal and vertical axes of the visualization chart, and then the content that students need to learn in a targeted manner can be found from the teaching audio / video.

[0118] In the implementation process of the above embodiment: first, the target time when the student's learning status is not good is determined from the visual chart, and the content that the student needs to learn in a targeted manner is determined from the teaching audio / video according to the target time. In this way, the teacher can more accurately identify the student's difficulties in specific knowledge points and provide timely and targeted tutoring, thereby improving the student's learning effect. At the same time, this method also helps teachers better understand the student's learning process, optimize teaching strategies, and improve teaching quality.

[0119] Optionally, in an embodiment of the present application, the teaching auxiliary information includes information representing the learning status of the class; the teaching auxiliary information is obtained according to the learning status corresponding to the learning activity, including: generating information representing the learning status of the class according to the learning status of multiple students participating in the course, and sending the information representing the learning status of the class to the teacher terminal.

[0120] According to the learning status of multiple students participating in the course, information representing the learning status of the class is generated. The information representing the learning status of the class can be information generated after integrating the learning status of multiple students, such as information representing the learning status of the class obtained by summarizing the visual charts of the learning status of multiple students, or directly summarizing the data corresponding to the learning status of the students as information representing the learning status of the class. The above method does not require calculation or processing of the learning status, and the speed of generating information representing the learning status of the class is faster, and the teacher can see the most original data on the teacher terminal.

[0121] Among them, summarizing the visualization charts of the learning status of multiple students refers to displaying the visualization charts of the learning status of multiple students using a general chart. For example, the visualization chart of each student's learning status is arranged vertically or horizontally, and the summarized visualization chart is used as information representing the learning status of the class. The real-time learning status of multiple students can be uniformly displayed through the summarized visualization chart.

[0122] The information representing the learning status of the class may also be information in a preset form generated after calculating or processing the learning status of multiple students. The information in the preset form may include at least one of a dot matrix formed by the learning status of multiple students or a data summary of the learning status of multiple students.

[0123] The dot matrix formed by the learning status of multiple students includes multiple points. The points can be arranged according to the class number or seat order. Each point represents a student. The points use different colors to display the student's learning status. For example, when the student's learning status is poor (concentration is lower than the threshold or the emotional state is not good), the color of the point corresponding to the student is red; when the student's learning status is good (concentration is higher than the threshold or the emotional state is good), the color of the point corresponding to the student is green (the color is only an example and can be replaced by any other color). Or different facial expressions can be used at the points to display the student's learning status, including facial expressions corresponding to pleasure, confusion, tension, etc. Or each point can also use different numbers to display the student's learning status. For example, the higher the value, the better the student's learning status; the lower the value, the worse the student's learning status. The dot matrix can more intuitively and eye-catchingly view the information representing the class learning status and improve the efficiency of the response.

[0124] The information in the preset form can also be a data summary of the learning status of multiple students, such as the average heart rate of all students participating in the same course, the average heart rate change, or other physiological indicators. This data summary can reflect the learning status of the students as a whole, and can therefore be used as information to characterize the learning status of the class.

[0125] The information representing the class learning status is sent to the teacher's terminal in real time. The teacher's terminal can be a mobile phone, tablet computer, laptop, smart watch or electronic device on the podium. The teacher can view the information representing the class learning status in real time, understand the real-time status changes of students in the course, and make timely teaching adjustments.

[0126] Adjusting the teaching pace includes: If the learning shows that the students' concentration has decreased, the teacher can introduce interactive sessions, ask questions or provide short breaks to restore the students' concentration. Adjusting the teaching pace also includes adjusting the difficulty of the content: If multiple students are generally nervous or confused about a complex concept, the teacher can slow down the teaching speed and provide more explanations or examples.

[0127] Furthermore, by tracking visualization charts over a long period of time and analyzing the correlation between learning status and course time in visualization charts, teachers can identify student performance trends in specific types of learning activities or target courses, thereby making longer-term teaching plans. For example, if multiple students participating in the same course show collective high concentration or good learning status under a certain teaching mode, then it is possible that students have a better ability to accept this teaching mode, and this type of teaching mode can be added to help students conduct learning activities in a better learning state and improve learning outcomes.

[0128] In an optional embodiment, after generating the information representing the class learning status, the information representing the class learning status can be analyzed to obtain some teaching suggestions. For example, if the information representing the class learning status shows that the students' overall concentration is reduced, some interesting links can be designed to attract the students' attention and improve their concentration. When the information representing the class learning status is sent to the teacher's terminal, the above suggestions and the designed links can be sent to the teacher's terminal together. If the information representing the class learning status shows that a classmate's learning status is obviously abnormal, a special attention prompt can also be sent to the teacher's terminal.

[0129] During the implementation of the above embodiment: information representing the class learning status is sent to the teacher's terminal, and the teacher can view the learning status of the students in the class and understand the real-time status changes of the students in the course. By sending the information representing the class learning status to the teacher's terminal to establish a good feedback loop, the teacher can adjust the teaching rhythm according to the students' learning status. The students' performance in turn affects the adjustment of teaching content and methods, which helps teachers optimize teaching strategies and improve teaching quality.

[0130] Optionally, in the embodiment of the present application, after collecting physiological data of students in learning activities through wearable devices, the method further includes:

[0131] According to the physiological data of students participating in the same course, the physiological index change data of the students participating in the same course are obtained; the physiological index change data represent the physiological data change characteristics of students in the learning activities. Students participating in the same course can be students in the same class or students who take the same course. Physiological data change characteristics include heart rate change rate, heart rate variability (HRV), breathing depth and rhythm changes, amplitude changes of skin electrodermal activity (EDA), body temperature changes, blood pressure fluctuations, blood oxygen saturation changes, exercise volume changes, etc. Among them, heart rate variability refers to the change in the time interval between successive heartbeats, reflecting the activity of the autonomic nervous system. Changes in the depth and rhythm of breathing (respiratory frequency) can reflect the emotional state and stress level of an individual. Changes in the depth and rhythm of breathing (respiratory frequency) can reflect the emotional state and stress level of an individual. Continuous changes in body temperature may reflect the metabolic activity of an individual or the influence of the external environment. Dynamic changes in blood pressure can reflect the state and stress response of the cardiovascular system. Exercise volume data collected by sensors such as accelerometers can reflect the physical activity level of an individual.

[0132] The physiological index change data of students in the same environment and taking the same course have some similarities. For example, if multiple students in the same course have similar physiological index changes at the same time (such as a sudden increase in heart rate), this may indicate a common external stimulus (such as an interesting question raised by the teacher) rather than individual students being distracted. On the contrary, if a student's physiological index is significantly different from that of his classmates, this may suggest that the student is distracted or absent-minded.

[0133] Therefore, the target students can be identified from the students who participate in the same course according to the physiological index change data of the students who participate in the same course; the target students are students whose learning status is abnormal. For example, the physiological index change data of the target students are compared with the average value of the physiological index change data of the students who participate in the same course. If the physiological index change data of the target students is obviously higher or lower than the average value of the physiological index change data, it means that the learning status of the target students is abnormal.

[0134] Alternatively, an indicator change chart can be drawn based on the physiological indicator change data of students participating in the same course. The indicator change chart can intuitively present the physiological indicator change data of each student, and the target students whose physiological indicator change data are obviously different from those of other students can be identified from the indicator change chart.

[0135] In the implementation process of the above embodiment: according to the physiological index change data of students participating in the same course, the target students (i.e. students with abnormal learning status) are determined, and teachers can provide personalized guidance and support for these students; and by analyzing the changes in the physiological indicators of the group, the impact of common external stimuli on students can be identified. This group behavior analysis helps to understand the classroom dynamics, help teachers adjust teaching strategies, and use group reactions to design more attractive teaching activities.

[0136] Optionally, in the embodiment of the present application, after analyzing the physiological data to determine the learning status of the student in the learning activity, the method further includes:

[0137] When it is determined through physiological data that the student's emotional state represents fear, it is judged whether the student is in a dangerous state based on at least one of the acceleration data, gyroscope data, audio data and video data collected by the wearable device; if so, an alarm message is generated and sent to a preset terminal in real time; the alarm message includes location information.

[0138] For example, the collected physiological data (such as heart rate, heart rate variability (HRV) and / or electrodermal activity (EDA)) can be input into the state analysis model, which includes an emotion analysis model. The emotion analysis model can output the category of emotion based on the input physiological data. It is determined that the student's emotional state represents fear, and further determines whether the student is in a dangerous state. Dangerous states include falling, being beaten and insulted, being bullied or kidnapped, etc.

[0139] For the application of acceleration data and gyroscope data: For example, through the built-in accelerometer and gyroscope of the wearable device, the acceleration data and gyroscope data of the student are monitored in real time to determine the student's movement status. Set the acceleration and angular velocity thresholds for falls. When a student suddenly experiences abnormal acceleration changes (such as a sudden increase or decrease in acceleration data) and angular velocity changes (such as rapid rotation or tilt), the device will determine that the student may fall. If a fall is detected, the device should immediately issue an alarm and upload the data to the background system so that school administrators can respond in time. In addition to falls, other abnormal movement states of students, such as being dragged or forced to move, can also be monitored through acceleration data and gyroscope data. By analyzing the fluctuation pattern of the data, these abnormal behaviors can be identified, alarm information can be generated, and alarms can be triggered.

[0140] For the application of audio data: Using audio recognition technology, wearable devices can monitor and identify the audio content around students in real time. For example, set a keyword library, including negative words such as insults and threats. When the device detects these keywords, an alarm will be triggered. The system can also further analyze the tone, volume and other characteristics of the audio to more accurately determine whether there is bullying or insulting behavior. Abnormal sound detection can also be performed. By analyzing the spectrum and waveform characteristics of the audio data, abnormal sounds such as screams and fighting sounds can be identified. These abnormal sounds are usually associated with violent incidents or dangerous situations, so the device should immediately generate an alarm message when these sounds are detected.

[0141] Application of video data: Through the camera attached to the wearable device or linkage with other camera systems, students' video data can be collected in real time. Using image recognition and machine learning technology, students' behavior patterns can be identified, such as fighting, smoking, climbing over walls, etc. If these abnormal behaviors are detected by the device based on the images in the video data, an alarm will be automatically triggered, and the video data will be uploaded to the background system for further analysis. In addition to behavioral recognition, facial expressions and postures in the video can also be analyzed: In addition to behavioral recognition, to determine whether they are in a dangerous state. For example, when a student is detected to show negative emotions such as fear, tension or pain, the device can issue an early warning. At the same time, by analyzing the student's posture (such as whether it is bound, blocked, etc.), it can also be determined whether they are in dangerous situations such as kidnapping.

[0142] It is understandable that when judging whether a student is in a dangerous state, one or more of acceleration data, gyroscope data, audio data and video data can be used for analysis. If a solution using multiple data includes fusing and comprehensively analyzing acceleration data, gyroscope data, audio data and video data, the accuracy of the judgment can be improved. By establishing a complex algorithm model, information from multiple data sources can be comprehensively considered to more accurately judge whether a student is in a dangerous state.

[0143] If a student is judged to be in danger, an alarm message is generated and sent to the preset terminal in real time; the alarm message includes location information. School administrators can view the student's location information and status information in real time through the background system so that they can respond quickly and take measures. The alarm message can also be pushed to the student's emergency contact or school security department so that they can intervene and provide help in time.

[0144] In the implementation process of the above embodiment: first, the emotional state is judged, and then the various data types collected by wearable devices are combined with analysis technology to achieve real-time monitoring and early warning of students' dangerous states, which helps schools take timely measures to protect students' safety and reduce potential safety risks.

[0145] Optionally, in the embodiment of the present application, the physiological data includes at least one of heart rate data, respiratory rate, skin electrode activity, body temperature, blood pressure data, blood oxygen saturation and exercise volume; collecting students' physiological data in learning activities through wearable devices includes: collecting students' physiological data in learning activities through wearable devices according to the course type corresponding to the learning activity; the course type includes at least one of cultural course type, physical education course type, art course type and practical course type. Among them, practical courses include experimental courses, manual courses and / or off-campus practical courses, and art courses include music courses, dance courses, broadcasting courses, performance courses, painting courses and / or sculpture courses, etc.

[0146] The latest course schedule information can be obtained regularly through the API interface by connecting to the school's academic affairs system, or students can be allowed to manually import or synchronize the course schedule through the wearable device application. The course schedule data is parsed to extract key information such as the type of each course (such as cultural courses, physical education courses, art courses, practical courses, etc.), time, location, etc. The parsed course schedule information can be stored in the local storage of the wearable device so that it can be used even when there is no network connection.

[0147] According to the hardware capabilities of the wearable device, set the basic collection frequency and parameters of each collection module (such as accelerometer, gyroscope, heart rate sensor, blood oxygen sensor, etc.). According to the course type in the curriculum, mobilize the collection module of the wearable device and control the collection frequency to collect students' physiological data during learning activities. For example, if the course type is a cultural course, and cultural courses mainly focus on mental activities, then the frequency of physical activity collection can be reduced, or the collection of physical activity data can be turned off to reduce the opening of too many functional modules, which causes excessive power consumption.

[0148] For example, if the course type is physical education, physical education requires a lot of physical activity, so the frequency of exercise volume collection should be increased, including steps, speed, acceleration, calories consumed, etc. At the same time, physiological data collection such as heart rate, blood oxygen, and body temperature should be enabled to monitor students' exercise intensity and health status. The GPS module can also be enabled to record students' exercise trajectory.

[0149] For other types of courses, such as art classes and practical classes, corresponding collection parameters (including collection modules and collection frequencies) can be configured to collect students' physiological data. For example, for art classes, considering that such courses usually involve creativity, imagination, and emotional expression, and may also include certain physical activities, such as music, painting, dancing, and performance, the collection of heart rate, blood oxygen saturation, exercise volume, audio and video data can be enabled to more comprehensively understand students' emotional state, participation, and performance in art classes. It can be understood that the above-mentioned different types of courses can set the collection modules and collection frequencies that need to be mobilized according to needs, and this application does not limit the specific collection mode of each course.

[0150] In the implementation process of the above embodiment: dynamically adjusting the collection module configuration of the wearable device according to the course type to realize the intelligent collection and management of students' physiological data can not only improve the pertinence and accuracy of physiological data collection, but also effectively save power and extend the use time of the wearable device.

[0151] In an optional embodiment, data collection may cover the entire learning activity process, including class time, breaks, and possible after-school activities. After data collection, the collected data can be divided into different time periods according to the curriculum, such as class time, breaks, lunch time, after-school activities, etc. According to the time period and the type of learning activity, the adaptation rules of heart rate fluctuations are formulated. For example, during class time, if the heart rate fluctuation is within a certain range (such as 60-100 times / minute), it is considered normal; if it exceeds this range, it may indicate that the learning state is not good. During the break time, the heart rate fluctuation may increase significantly, and the judgment criteria for heart rate fluctuations should be relaxed at this time.

[0152] In addition, the adaptation rules of heart rate fluctuations can be dynamically adjusted according to the actual situation of students and changes in their learning status. For example, if a student shows particularly high concentration and learning efficiency in a certain class, the judgment criteria of heart rate fluctuations can be adjusted accordingly. At the same time, the system should provide a feedback mechanism so that teachers and students can adjust and optimize the adaptation rules of heart rate fluctuations when necessary.

[0153] Furthermore, the collected physiological data can also be preprocessed, including data filtering and noise elimination. For example, the collected physiological data can be filtered to remove high-frequency noise and outliers. The filtering method can be selected according to the characteristics and needs of the data, such as using sliding average filtering, Kalman filtering, etc. The filtered data should be smoother and can more accurately reflect the physiological state of the students. Noise data caused by external factors can also be identified and eliminated. External factors may include recess activities, mood swings, equipment failures, etc. Noise elimination can be achieved by setting reasonable thresholds, using data cleaning algorithms, or combining contextual information.

[0154] Optionally, in the embodiment of the present application, analyzing the physiological data to determine the learning status of the student in the learning activity includes:

[0155] In the case of a cultural course, if the amount of exercise in the physiological data exceeds the preset threshold, the learning state is determined to be poor. A reasonable exercise threshold can be set as the preset threshold based on relevant research and the student's daily activity level. When the student's activity exceeds the preset threshold, it is considered that the student may be distracted due to excessive activity and have a poor learning state.

[0156] Alternatively, if the physiological data represents the student's static state, and the static state exceeds a preset time, the learning state is determined to be poor. Also based on research and students' daily behavior patterns, a threshold for static time is set as the preset time. When a student remains static for more than this time, it is considered that the student may be in a poor learning state due to lack of activity or interest.

[0157] In an optional embodiment, when it is detected that the student's activity suddenly increases and exceeds a preset exercise threshold, a warning is triggered, indicating that the student's learning state may be affected by the activity and is in a poor state. At this time, the student's heart rate changes, activity type and other information can be further analyzed to confirm whether the poor learning state is indeed caused by excessive activity.

[0158] When it is detected that the student remains still for a period of time exceeding the preset stillness time threshold, a warning is also triggered. In this case, the student's heart rate changes, eye movements and other information can be further analyzed to determine whether the student is distracted, tired or bored.

[0159] After determining that the student's learning state is not good, a warning is issued through a wearable device or teacher's terminal to alert students and teachers. Provide personalized intervention suggestions based on the reasons for the student's poor learning state. For example, for students whose learning state is poor due to excessive activity, it is recommended that they take a short rest or relaxation activities; for students whose learning state is poor due to long-term stillness, they may be in a sleepy state, and reminders can be issued through wearable devices, suggesting that they stand appropriately, look into the distance or other activities to help them refresh their minds. Of course, more suggestions can be given, such as reminding them to go to bed as soon as possible at a preset time in the evening (such as 10 or 11 o'clock in the evening), or giving some sleep suggestions by monitoring sleep to improve sleepiness in class.

[0160] In the implementation process of the above embodiment: the learning status of students in the process of learning cultural courses is evaluated by monitoring their physiological data, and the physiological data of students, such as heart rate, activity level, etc., are collected and analyzed, and it is judged whether the learning status of students is in a bad state based on this. The learning status of students in the process of learning cultural courses can be understood more accurately, so that effective intervention measures can be taken in time to improve students' learning efficiency and effect.

[0161] Optionally, in an embodiment of the present application, the learning status includes emotional state or concentration. After analyzing the physiological data and determining the student's learning status in the learning activity, the method also includes: generating status information based on the student's learning status in the learning activity; and sending the status information to the student's parent terminal.

[0162] Before collecting and sending student learning status information, ensure compliance with data privacy regulations and obtain consent from parents and students. Status information such as charts, line graphs, patterns, color coding or concise text descriptions, so that parents can quickly grasp their children's learning status. Provide multiple channels for sending status information, such as mobile applications, emails or text messages, to ensure that parents can easily receive information.

[0163] Taking text description as an example, the status information may include timestamp, course name, teacher feedback, recommended actions, etc. to provide a comprehensive snapshot of the learning status. Exemplarily, the status information may be a description of the emotional state (e.g., "Student A showed anxiety in the 10 a.m. math class. It is recommended that parents discuss the experience in class with Student A and encourage them to ask questions in class. The teacher plans to provide more one-on-one guidance to Student A in the next class.") and a concentration score (e.g., "Student B scored 7 out of 10 in the 2 p.m. history class. He performed very well and can be encouraged more.").

[0164] You can also draw corresponding charts for more intuitive viewing, such as using emotional icons, emotional wheels (circular charts that distribute different emotional states (such as happiness, sadness, anxiety, excitement, etc.) on the circumference, and use colors or pointers to indicate the current emotions according to the students' emotional states), heat maps (use color changes to indicate students' concentration or emotional states in different time periods. The "hotter" the color (such as red), the stronger the concentration or emotional state), line graphs, pie charts (showing the distribution ratio of students' different emotional states in a day or a week, with each sector representing an emotion) to indicate pleasure, or use progress bars to indicate concentration. Parents can freely choose the form of the above status information, and can also provide parents with daily, weekly or monthly reports to summarize students' learning status in different time periods.

[0165] As an implementation method, the parent terminal can also install a parent-side application that provides richer functions, such as viewing historical status information, receiving notifications, communicating with teachers, etc. A feedback loop can be established through the application to encourage parents to provide feedback, teachers adjust teaching strategies based on the feedback, and then update the status information again.

[0166] In the implementation process of the above embodiment: parents can understand their children's performance in school in real time, enhance communication and cooperation with teachers. By establishing a more open, transparent and interactive education environment, parents can become active participants in their children's education and jointly promote their children's learning and growth.

[0167] Optionally, in an embodiment of the present application, status information is generated according to the learning status of the student in the learning activity, including: generating status information according to the course time and learning status corresponding to the learning activity; the status information includes learning status corresponding to different times and / or learning status corresponding to different courses.

[0168] Continuing with the above embodiment, the status information can be a bar graph or a line graph, the horizontal axis can be different times or different courses, and the vertical axis is the learning status corresponding to the time or the course; if the time is a time point, the learning status is the emotional state or concentration corresponding to the time point; if the time is a time period, the learning status is the mean of the emotional state or the mean of the concentration in the time period; wherein the mean can also be replaced by other data that can characterize the learning status of the time period, such as the emotional state or concentration that appears most frequently in the time period.

[0169] As an implementation method, the status information may first display the learning status of a time period (eg, a certain day), and further clicking on the time period may display the learning status of different courses within the time period (a day).

[0170] In the above implementation process, parents can view students' learning status at different times and / or different courses in real time, pay attention to changes in students' learning status in a timely manner, provide help and support, and understand students' status at each time or in each class more clearly and conveniently with finer granularity, which helps to evaluate changing trends in students' learning habits, stress levels and health conditions.

[0171] See also Figure 2 The structural diagram of the teaching evaluation device provided in the embodiment of the present application is shown; the embodiment of the present application provides a teaching evaluation device 200, including:

[0172] The collection module 210 is used to collect physiological data of students during learning activities through wearable devices;

[0173] An analysis module 220 is used to analyze the physiological data to determine the learning status of the student during the learning activity;

[0174] The evaluation module 230 is used to obtain teaching auxiliary information according to the learning status corresponding to the learning activity.

[0175] Optionally, in an embodiment of the present application, the teaching evaluation device 200, the physiological data includes heart rate data; the analysis module 220 is used to extract heart rate characteristics within the learning activity from the heart rate data; the heart rate characteristics include heart rate statistics, heart rate change rate, heart rate variability and / or heart rate spectrum characteristics within the course time corresponding to the learning activity; based on the heart rate characteristics and a pre-trained state analysis model, the learning state in the learning activity is obtained; the learning state includes emotional state or concentration.

[0176] Optionally, in an embodiment of the present application, the teaching evaluation device 200, the physiological data also includes derived data; the derived data includes at least one of respiratory rate, skin electrode activity, body temperature, blood pressure data, blood oxygen saturation and exercise volume; the analysis module 220 is specifically used to extract derived features in learning activities from the derived data; the derived features include statistical values ​​of derived data in learning activities, the correlation between derived data and heart rate data, or the correlation between derived data; the heart rate features and the derived features are combined to generate physiological features; the physiological features are input into a pre-trained state analysis model to obtain the learning state in the learning activity.

[0177] Optionally, in an embodiment of the present application, the teaching evaluation device 200 also includes a model training module, which is used to obtain sample heart rate data and sample derived data under different learning states; obtain sample heart rate features based on the sample heart rate data; and obtain sample derived features based on the sample derived data; the sample derived features include statistical values ​​of sample derived data under different learning states, the correlation between sample derived data and sample heart rate data, or the correlation between sample derived data; the sample heart rate features and sample derived features under the same learning state are combined, and a state label corresponding to the learning state is added to obtain a sample feature vector; the state label includes an emotion label or a concentration label; the machine learning model is trained using the sample feature vector to obtain a state analysis model; the state analysis model includes an emotion analysis model or a concentration model.

[0178] Optionally, in an embodiment of the present application, the teaching evaluation device 200, the teaching auxiliary information includes the student's liking for the target course; the evaluation module 230 is specifically used to obtain the student's learning status corresponding to multiple learning activities for the target course, and determine the student's liking for the course based on the proportion of times the learning status is well represented in the learning status corresponding to the multiple learning activities.

[0179] Optionally, in an embodiment of the present application, the teaching evaluation device 200, the teaching auxiliary information includes the students' learning effects on the courses corresponding to the learning activities; the evaluation module 230 is specifically used to determine the students' learning effects on the courses corresponding to the learning activities according to the students' learning status corresponding to the learning activities, and provide personalized homework guidance according to the learning effects; wherein the learning status and the learning effects are positively correlated.

[0180] Optionally, in an embodiment of the present application, the teaching evaluation device 200, the teaching auxiliary information includes a visualization chart; the evaluation module 230 is specifically used to draw a visualization chart based on the real-time acquisition of the learning status of students in different learning activities; the axes of the visualization chart are respectively the time and learning status corresponding to the learning activity; the visualization chart is used as a reference object for adjusting the teaching rhythm.

[0181] Optionally, in an embodiment of the present application, the teaching auxiliary information includes content that students need targeted tutoring; the teaching evaluation device 200, the evaluation module 230, is specifically used to determine the target time when the student's learning status is poor according to the student's corresponding learning status in different learning activities, and determine the content that students need targeted tutoring from the teaching audio / video according to the target time; the teaching audio / video is obtained by recording the teacher's teaching process.

[0182] Optionally, in an embodiment of the present application, the teaching evaluation device 200 further includes: a teaching rhythm adjustment module, which is used to send the visual chart to the teacher terminal in real time, so that the teacher can dynamically adjust the teaching rhythm according to the status of the students in the visual chart.

[0183] Optionally, in an embodiment of the present application, the teaching evaluation device 200 also includes: a group analysis module, which is used to obtain physiological indicator change data of students participating in the same course based on the physiological data of students participating in the same course; the physiological indicator change data represents the physiological data change characteristics of students in learning activities; based on the physiological indicator change data of students participating in the same course, target students are determined from students participating in the same course; the target students are students whose learning status characteristics are abnormal.

[0184] Optionally, in an embodiment of the present application, the teaching evaluation device 200 further includes: a danger detection module, which is used to determine whether the student is in a dangerous state based on at least one of the acceleration data, gyroscope data, audio data and video data collected by the wearable device when it is determined through physiological data that the student's emotional state represents fear; if so, an alarm message is generated and sent to a preset terminal in real time; the alarm message includes location information.

[0185] Optionally, in an embodiment of the present application, the teaching evaluation device 200, the physiological data includes at least one of heart rate data, respiratory rate, skin electrode activity, body temperature, blood pressure data, blood oxygen saturation and exercise volume; the collection module 210 is specifically used to collect physiological data of students in learning activities through wearable devices according to the course type corresponding to the learning activity; the course type includes at least one of cultural course type, physical education course type, art course type and practical course type.

[0186] Optionally, in an embodiment of the present application, the teaching evaluation device 200, the analysis module 220, is also used to, when the course type is a cultural course type, determine that the learning status is poor if the amount of exercise in the physiological data exceeds a preset threshold; or, if the physiological data represents the student's static state and the static state exceeds a preset duration, determine that the learning status is poor.

[0187] Optionally, in an embodiment of the present application, the learning status includes an emotional state or concentration, and the teaching evaluation device 200 further includes a parent interaction module for generating status information according to the student's learning status in the learning activity; and sending the status information to the student's parent terminal.

[0188] Optionally, in an embodiment of the present application, the teaching evaluation device 200, the parent interaction module, is specifically used to generate the status information according to the course time and the learning status corresponding to the learning activity; the status information includes the learning status corresponding to different times and / or the learning status corresponding to different courses.

[0189] It should be understood that the device corresponds to the above-mentioned teaching evaluation method embodiment and can execute the various steps involved in the above-mentioned method embodiment. The specific functions of the device can be found in the above description. To avoid repetition, the detailed description is appropriately omitted here. The device includes at least one software function module that can be stored in a memory in the form of software or firmware or solidified in the operating system (OS) of the device.

[0190] See also Figure 3 The electronic device 300 provided in the embodiment of the present application includes: a processor 310 and a memory 320, wherein the memory 320 stores machine-readable instructions executable by the processor 310, and when the machine-readable instructions are executed by the processor 310, the above method is executed.

[0191] Figure 3 The components shown in can be implemented by hardware, software or a combination thereof. The electronic device 300 may be a physical device, such as a wearable device, a server, a PC, etc., or a virtual device, such as a virtual machine, a virtualized container, etc. Moreover, the electronic device 300 is not limited to a single device, but may also be a combination of multiple devices or a cluster composed of a large number of devices. Exemplarily, if it is a single device, the electronic device may be a wearable device, and after the wearable device collects physiological data, the learning status may be analyzed locally and teaching auxiliary information may be obtained; the electronic device may also be multiple devices, including wearable devices and servers; the wearable device may be used to collect the physiological data of students in learning activities, and the collected physiological data may be sent to the server for processing, including determining the learning status of the learner in the learning activity, and providing teaching auxiliary information according to the learning status activity.

[0192] An embodiment of the present application further provides a storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the above method is executed.

[0193] Among them, the storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable red-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, disk or optical disk.

[0194] In several embodiments provided in the embodiments of the present application, it should be understood that the disclosed devices and methods can also be implemented in other ways. The device embodiments described above are merely schematic. For example, the flowcharts and block diagrams in the accompanying drawings show the possible architecture, functions and operations of the devices, methods and computer program products according to the multiple embodiments of the present application. In this regard, each box in the flowchart or block diagram can represent a module, a program segment or a part of the code, and a module, a program segment or a part of the code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order than the order marked in the accompanying drawings. For example, two consecutive boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flowchart, and; the combination of boxes in the block diagram and / or flowchart can be implemented with a dedicated hardware-based system that performs a specified function or action, or can be implemented with a combination of dedicated hardware and computer instructions.

[0195] In addition, the functional modules in each embodiment of the present application can be integrated together to form an independent part, or each module can exist separately, or two or more modules can be integrated to form an independent part.

[0196] The above description is only an optional implementation manner of the embodiments of the present application, but the protection scope of the embodiments of the present application is not limited thereto. Any technician familiar with the technical field can easily think of changes or replacements within the technical scope disclosed in the embodiments of the present application, which should be covered within the protection scope of the embodiments of the present application.

Claims

1. A teaching evaluation method, characterized in that: include: Collect students’ physiological data during learning activities through wearable devices; Analyzing the physiological data to determine the learning status of the student in the learning activity; According to the learning status corresponding to the learning activity, teaching auxiliary information is obtained.

2. The method according to claim 1, characterized in that The physiological data includes heart rate data; and analyzing the physiological data to determine the learning status of the student in the learning activity includes: Extracting the heart rate features in the learning activity from the heart rate data; the heart rate features include heart rate statistics, heart rate change rate, heart rate variability and / or heart rate spectrum features during the course time corresponding to the learning activity; The learning state in the learning activity is obtained according to the heart rate characteristics and a pre-trained state analysis model; the learning state includes an emotional state or a concentration level.

3. The method according to claim 2, characterized in that The physiological data also includes derived data; the derived data includes at least one of respiratory rate, skin electrode activity, body temperature, blood pressure data, blood oxygen saturation and exercise volume; the learning state in the learning activity is obtained according to the heart rate characteristics and the pre-trained state analysis model, including: Extracting derived features in the learning activity from the derived data; the derived features include derived data statistics in the learning activity, associations between the derived data and heart rate data, or associations between the derived data; Combining the heart rate feature and the derived feature to generate a physiological feature; The physiological characteristics are input into the pre-trained state analysis model to obtain the learning state in the learning activity.

4. The method according to claim 3, characterized in that Before inputting the physiological characteristics into the pre-trained state analysis model to obtain the learning state in the learning activity, the method further includes: Obtain sample heart rate data and sample derivative data under different learning states; Obtain sample heart rate features according to the sample heart rate data; and obtain sample derived features according to the sample derived data; the sample derived features include statistical values ​​of sample derived data under different learning states, associations between the sample derived data and sample heart rate data, or associations between the sample derived data; Combining the sample heart rate feature and the sample derived feature in the same learning state, and adding a state label corresponding to the learning state, to obtain a sample feature vector; the state label includes an emotion label or a concentration label; The sample feature vector is used to train a machine learning model to obtain the state analysis model; the state analysis model includes a sentiment analysis model or a concentration model.

5. The method according to claim 1, characterized in that The teaching auxiliary information includes the student's preference for the target course; the obtaining of the teaching auxiliary information according to the learning status corresponding to the learning activity includes: The learning status corresponding to multiple learning activities of the student for the target course is obtained, and the student's liking for the course is determined according to the proportion of times in which the learning status is well characterized among the learning status corresponding to the multiple learning activities.

6. The method according to claim 1, characterized in that The teaching auxiliary information includes the learning effect of the student on the course corresponding to the learning activity; the obtaining of the teaching auxiliary information according to the learning status corresponding to the learning activity includes: According to the learning status of the student corresponding to the learning activity, the learning effect of the student on the course corresponding to the learning activity is determined, and personalized homework guidance is performed according to the learning effect; wherein the learning status and the learning effect are positively correlated.

7. The method according to claim 1, characterized in that The teaching auxiliary information includes a visual chart; the obtaining of the teaching auxiliary information according to the learning state corresponding to the learning activity includes: A visualization chart is drawn based on the learning status of the students in different learning activities obtained in real time; the axes of the visualization chart are respectively the time and the learning status corresponding to the learning activities; the visualization chart is used as a reference object for adjusting the teaching rhythm.

8. The method according to claim 1, characterized in that: The teaching auxiliary information includes the content that the student needs to learn in a targeted manner; the obtaining of the teaching auxiliary information according to the learning status corresponding to the learning activity includes: According to the corresponding learning status of the student in different learning activities, the target time when the student's learning status is poor is determined, and according to the target time, the content that the student needs targeted tutoring is determined from the teaching audio / video; the teaching audio / video is obtained by recording the teacher's teaching process.

9. The method according to claim 1, characterized in that: The teaching auxiliary information includes information representing the learning status of the class; the obtaining of the teaching auxiliary information according to the learning status corresponding to the learning activity includes: According to the learning status of the plurality of students participating in the course, information representing the learning status of the class is generated, and the information representing the learning status of the class is sent to the teacher terminal.

10. The method according to claim 1, characterized in that After collecting physiological data of students during learning activities through wearable devices, the method further includes: Obtaining physiological index change data of students participating in the same course according to the physiological data of students participating in the same course; the physiological index change data represents the physiological data change characteristics of the students in the learning activity; According to the physiological index change data of the students participating in the same course, a target student is determined from the students participating in the same course; the target student is a student with abnormal learning status characteristics.

11. The method according to claim 1, characterized in that After analyzing the physiological data to determine the learning status of the student in the learning activity, the method further includes: When it is determined through the physiological data that the student's emotional state represents fear, it is judged whether the student is in a dangerous state based on at least one of the acceleration data, gyroscope data, audio data and video data collected by the wearable device; if so, an alarm message is generated and sent to a preset terminal in real time; the alarm message includes location information.

12. The method according to claim 1, characterized in that The physiological data includes at least one of heart rate data, respiratory rate, skin electrode activity, body temperature, blood pressure data, blood oxygen saturation and exercise volume; The physiological data of students during learning activities collected through wearable devices include: According to the course type corresponding to the learning activity, collecting physiological data of students in the learning activity through wearable devices; The course type includes at least one of a cultural course type, a physical education course type, an art course type, and a practical course type.

13. The method according to claim 12, characterized in that Analyzing the physiological data to determine the learning status of the student in the learning activity includes: In the case where the course type is a cultural course type, if the amount of exercise in the physiological data exceeds a preset threshold, it is determined that the learning state is poor; Alternatively, if the physiological data represents the student's static state, and the static state exceeds a preset time, it is determined that the learning state is not good.

14. The method according to claim 1, characterized in that The learning state includes an emotional state or a concentration level. After analyzing the physiological data to determine the learning state of the student in the learning activity, the method further includes: Generate status information according to the learning status of the student in the learning activity; and send the status information to a parent terminal of the student.

15. The method according to claim 14, characterized in that The generating status information according to the learning status of the student in the learning activity includes: The status information is generated according to the course time corresponding to the learning activity and the learning status; the status information includes the learning status corresponding to different times and / or the learning status corresponding to different courses.

16. A computer program product, characterized in that The method comprises computer program instructions, and when the computer program instructions are executed by a processor, the method according to any one of claims 1 to 15 is executed.

17. An electronic device, characterized in that: include: A processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the method according to any one of claims 1 to 15 is executed.

18. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer program instructions, and when the computer program instructions are executed by a processor, the method according to any one of claims 1 to 15 is executed.

Images