Teaching data analysis method and system, medium and program product

Abstract

The invention relates to the technical field of data analysis, in particular to a teaching data analysis method and system, a medium and a program product, and the method comprises the steps: obtaining the movement information of a teacher in a preset teaching time period, and determining a teacher walking thermodynamic diagram of the teacher in the preset teaching time period based on the movement information; acquiring learning state information of the students in a preset teaching time period, and determining a learning state index of each student in the teaching time period based on the learning state information; determining a teaching coverage map based on the teacher walking thermodynamic diagram and the learning state index of each student; and determining a to-be-concerned area and to-be-concerned teaching content based on the teaching coverage map, and generating a teaching feedback suggestion based on the to-be-concerned area and the to-be-concerned teaching content, the teaching feedback suggestion being used for reminding the teacher to perform teaching adjustment in the next teaching time period. According to the invention, the integration degree of the AI teaching technology in the actual teaching process is improved, so that the teaching quality is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data analysis, in particular to a teaching data analysis method, an analysis system, a medium and a program product. BACKGROUND

[0002] Traditional teaching evaluation and analysis modes mostly rely on experience judgment, and have strong subjectivity, limited coverage, and delayed feedback, etc. Teachers often obtain student learning conditions through classroom observation, homework correction and stage examination, but it is difficult to fully capture the learning details and personalized needs of each student, especially in large class teaching scenarios, it is difficult to achieve targeted teaching intervention. Due to the limitations of traditional teaching evaluation and analysis modes, AI teaching technology emerges as the times require. The essence of the emergence of AI teaching technology is to compensate for the shortcomings of traditional teaching in precision, personalization and efficiency through technical means, so as to improve the teaching quality.

[0003] However, in actual implementation, many teachers may only use AI teaching technology at a shallow level, and fail to integrate it deeply into the whole teaching process, which may result in little improvement in teaching quality. SUMMARY

[0004] In order to improve the integration of AI teaching technology in the actual teaching process and improve the teaching quality, the present application provides a teaching data analysis method, an analysis system, a medium and a program product.

[0005] In a first aspect, the present application provides a teaching data analysis method, which adopts the following technical solution: A teaching data analysis method comprises: obtaining the movement information of a teacher in a preset teaching time period, and determining a teacher walking heat map of the teacher in the preset teaching time period based on the movement information; obtaining the learning state information of students in the preset teaching time period, and determining a learning state index of each student in the teaching time period based on the learning state information; determining a teaching coverage map based on the teacher walking heat map and the learning state index of each student; determining a region to be concerned and teaching content to be concerned based on the teaching coverage map, and generating a teaching feedback suggestion based on the region to be concerned and the teaching content to be concerned, the teaching feedback suggestion being used to remind the teacher to make teaching adjustment in the next teaching time period.

[0006] By employing the aforementioned technical solutions, and analyzing teachers' movement information during the teaching phase to generate a teacher movement heatmap, it is possible to intuitively present areas of high and low teacher attention. This helps teachers objectively recognize the limitations of their teaching behavior, proactively adjust their movement range, and ensure that every student receives basic attention. Furthermore, by analyzing students' learning status information during the teaching process and quantifying each student's classroom performance through a learning status index, it is possible to overcome the bottleneck of traditional classroom observation, which struggles to focus on the details of each student. By combining teacher movement and student learning status, it is possible to bridge the gap between teacher behavior and student feedback. Finally, by identifying areas and students requiring attention through a teaching coverage map, targeted teaching adjustments are generated, facilitating a data-driven teaching optimization loop. By directly transforming the results of AI teaching technology analysis into actionable teaching actions, it is possible to enhance the integration of AI teaching technology into the actual teaching process, thereby improving teaching quality.

[0007] In one possible implementation, determining the instructional coverage map based on the teacher movement heatmap and each student's learning status index includes: The teacher movement heatmap is partitioned based on the heat intensity value corresponding to each teaching location, resulting in multiple heat zones, each with a different mean heat intensity value. Based on the average thermal intensity of each thermal zone and the preset weight mapping relationship, the zone weight corresponding to each thermal zone is determined. The preset weight mapping relationship is the correspondence between the average thermal intensity and the zone weight. Based on the learning status index of each student in each heat zone and the corresponding zone weight, determine the zone coverage map corresponding to each heat zone; The teaching coverage map is obtained by integrating the zoning coverage maps of all thermal zones.

[0008] By adopting the above technical solution, the teaching movement heatmap is divided into regions based on the heat intensity values ​​corresponding to each teaching location. This facilitates the differentiation between areas of high teacher attention and areas of low attention. Different weights are assigned to different heatmap zones to avoid the one-size-fits-all problem caused by using fixed weights. By linking teacher attention levels with students' learning status indices, a connection between teaching and learning is established. By converting students' learning status indices within each heatmap zone into a visualized zone coverage map, abstract teaching and learning correlation data is transformed into intuitive and perceptible visual information. This allows relevant teachers to quickly locate core problem areas and problem types, thus providing clear visual guidance for subsequent teaching adjustments.

[0009] In one possible implementation, before determining the partition coverage map corresponding to a heat partition based on the learning state index of each student and the corresponding partition weight, the method further includes: Identify the learning status indices of any two students within the thermal zone, and determine the index difference between the learning status indices of any two students; If the difference between the learning status indices of any two students is higher than a preset difference threshold, the proportion of students with high and low status indices is determined based on the mean status index corresponding to the heat zone. The mean status index is determined by averaging the learning status indices of each student in the heat zone. The number of students to be removed based on the aforementioned percentage of students is determined as follows: the number of students removed based on their high status and the number of students removed based on their low status. Based on the number of high-state eliminations, the first elimination is carried out from students of the high-state index type whose learning state index is higher than the mean state, and the first elimination result is obtained. Based on the number of students to be removed from the low-state state, a second removal is carried out from students whose learning state index is lower than the mean state, and the second removal result is obtained. Based on the first elimination result and the second elimination result, update the learning status index of each student corresponding to the heat map partition.

[0010] By adopting the above technical solution, before determining the zoning coverage map corresponding to the thermal zoning, the student learning status index can be accurately cleaned and optimized by detecting the difference, classifying the types, removing anomalies, and updating the data. This avoids a large gap between the determined zoning coverage map and the actual teaching effect due to excessive dispersion of the student status index.

[0011] In one possible implementation, the method further includes: Acquire real-time learning status information for each student, identify the status characteristics of each student from each real-time learning status information, and identify the real-time status characteristic value of each status characteristic. The real-time state index of each student is determined based on the preset state feature weights and the real-time state feature values. Based on each student's real-time status index, a learning status feedback bar is generated for each student, and each learning status feedback bar is superimposed on the corresponding student position in the classroom teaching image to obtain a real-time classroom status feedback image.

[0012] By adopting the above technical solution, students' learning status is captured in real time, the status index is quantified, and the data is superimposed onto the classroom teaching image in the form of feedback bars. This allows relevant teachers to keep track of the fluctuations in students' status by viewing the classroom teaching image.

[0013] In one possible implementation, the method further includes: Identify the feedback bar features of each learning status feedback bar in the real-time classroom status feedback image, wherein the feedback bar features include the feedback bar growth rate and the feedback bar length; The learning status feedback bars that meet the preset anomaly judgment criteria are identified as anomaly feedback bars, and stars are superimposed on the anomaly feedback bars. The preset anomaly judgment criteria are: the growth rate of the feedback bar is lower than the preset growth rate threshold; or, the length of the feedback bar is lower than the preset length threshold. Identify the location of the starred student corresponding to each star within a preset observation period. When the distribution density of the starred student locations is greater than a preset density threshold and the number of starred students is higher than a preset number threshold, activate the one-click slicing mode to slice the classroom teaching segment corresponding to the preset observation period and obtain the starred slice teaching segment. Feedback will be provided on the starred slices of classroom teaching segments.

[0014] By adopting the above technical solution, the characteristics of each learning status feedback bar are detected in real time, and stars are superimposed on the learning status feedback bars that meet the preset abnormal judgment criteria. This facilitates accurate and intuitive identification of students with abnormal learning status. By statistically analyzing the number and position distribution of stars, the abnormality type can be automatically judged. After identifying the abnormality type as a group abnormality, a one-click slicing mode can be activated to automatically edit classroom teaching segments within a preset observation period. This eliminates the need for teachers to manually select segments after class, allowing for the rapid acquisition of core segments containing the "problem scenario," thereby reducing the time cost for teachers to conduct teaching reviews after class.

[0015] In one possible implementation, the method further includes: Obtain initial classroom teaching records and identify the teacher behavior occupancy rate and teacher-student interaction frequency corresponding to each teaching period in the initial classroom teaching records; Based on the teacher behavior occupancy rate and teacher-student interaction frequency corresponding to each teaching period, the time period classroom type and time period teaching content corresponding to each teaching period are determined, and the slice display format type of each teaching period is determined according to the time period classroom type corresponding to each teaching period. Based on the teaching content corresponding to each teaching period, the teaching focus markers and comprehensive learning status index of each teaching period are determined. Based on the teaching focus markers and the comprehensive learning status index, the slice display level type corresponding to each teaching period is determined. The comprehensive learning status index is determined by the learning status index of each student in the corresponding teaching period. Based on the slice display format type, slice display degree type, and teaching content of each teaching period, the teaching record of each teaching period is determined, and the teaching records of all teaching periods are integrated to obtain the target classroom teaching record corresponding to the initial classroom teaching record.

[0016] By adopting the above technical solution, and by analyzing the teacher behavior occupancy rate and teacher-student interaction frequency corresponding to each teaching moment, it is easy to divide the entire teaching process into time periods. Based on the actual teaching situation corresponding to each teaching period, the slice display format and slice display degree type are determined for each teaching period. Finally, the teaching records of each teaching period slice are integrated. This method of dividing and recording can avoid the redundancy problem of full recording. Through AI teaching technology, classroom records are transformed from passive storage into actively optimized structured resources, making the records more in line with teaching needs, thereby facilitating the efficient accumulation of teaching resources and the continuous improvement of teaching quality.

[0017] Secondly, this application provides an analysis system, which adopts the following technical solution: An analysis system comprising: At least one processor; Memory; At least one application, wherein the at least one application is stored in memory and configured to be executed by at least one processor, the at least one application being configured to: execute the teaching data analysis method described above.

[0018] Thirdly, this application provides a computer-readable storage medium, which adopts the following technical solution: A computer-readable storage medium includes: a computer program stored thereon that can be loaded by a processor and execute the above-described teaching data analysis method.

[0019] Fourthly, this application provides a computer program product, which adopts the following technical solution: A computer program product includes a computer program that, when executed by a processor, implements the above-described teaching data analysis method.

[0020] In summary, this application includes at least one of the following beneficial technical effects: By analyzing teachers' movement information during the teaching phase and generating a teacher movement heatmap, it is possible to intuitively present areas of high and low teacher attention. This helps teachers objectively recognize the limitations of their teaching behavior, proactively adjust their movement range, and ensure that every student receives basic attention. Furthermore, by analyzing students' learning status information during the teaching process and quantifying each student's classroom performance through a learning status index, it is possible to overcome the bottleneck of traditional classroom observation, which struggles to focus on the details of each student. By combining teacher movement and student learning status, it is possible to bridge the gap between teacher behavior and student feedback. Finally, by identifying areas and students requiring attention through a teaching coverage map, targeted teaching adjustments are generated, facilitating a data-driven teaching optimization loop. By directly transforming the results of AI teaching technology analysis into actionable teaching actions, it is possible to enhance the integration of AI teaching technology into the actual teaching process, thereby improving teaching quality.

[0021] By analyzing the teacher behavior occupancy rate and teacher-student interaction frequency corresponding to each teaching moment, it is easy to divide the entire teaching process into time periods. Based on the actual teaching situation of each teaching period, the type and degree of slice presentation for each teaching period are determined. Finally, the teaching records of each teaching period slice are integrated. This method of dividing and recording helps to avoid the redundancy problem of full recording. Through AI teaching technology, classroom records are transformed from passive storage into actively optimized structured resources, making the records more in line with teaching needs. This facilitates the efficient accumulation of teaching resources and the continuous improvement of teaching quality. Attached Figure Description

[0022] Figure 1 This is a flowchart illustrating a teaching data analysis method according to an embodiment of this application; Figure 2 This is a flowchart illustrating a process for determining target classroom teaching records in an embodiment of this application; Figure 3 This is a schematic diagram of the structure of an analysis system according to an embodiment of this application. Detailed Implementation

[0023] The following is in conjunction with the appendix Figures 1 to 3 This application will be described in further detail.

[0024] After reading this specification, those skilled in the art may make modifications to this embodiment without contributing any inventive step, but such modifications are protected by patent law as long as they fall within the scope of the claims of this application.

[0025] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0026] It should be noted that, in the optional embodiments of this application, the data related to object information, when applied to specific products or technologies, requires the permission or consent of the object. Furthermore, the collection, use, and processing of this data must comply with the relevant laws, regulations, and standards of the relevant countries and regions. In other words, if the embodiments of this application involve data related to an object, it must be obtained with the object's authorization and consent, the authorization and consent of relevant departments, and in accordance with the relevant laws, regulations, and standards of the country and region. If the embodiments involve personal information, the acquisition of all personal information requires the individual's consent. If sensitive information is involved, the separate consent of the information subject is required. The embodiments also need to be implemented with the object's authorization and consent.

[0027] Specifically, this application provides a teaching data analysis method executed by an analysis system. This analysis system can be a server or a terminal device. The server can be a standalone physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing cloud computing services. The terminal device can be a smartphone, tablet, laptop, desktop computer, etc., but is not limited to these. The terminal device and the server can be directly or indirectly connected via wired or wireless communication, and this application does not impose any limitations on this connection.

[0028] refer to Figure 1 , Figure 1 This is a flowchart illustrating a teaching data analysis method according to an embodiment of this application. The method includes steps S110-S140, wherein: Step S110: Obtain the movement information of teachers during the preset teaching time period, and determine the teacher movement heat map during the preset teaching time period based on the movement information.

[0029] Specifically, the preset teaching time period is a period of time after the start of class. The duration of the preset teaching time period can be 10 minutes or 15 minutes, and the entire course can include multiple preset teaching time periods. The teacher's movement information within the preset teaching time period can be collected by image acquisition devices installed on the classroom ceiling, front and back walls, etc., and uploaded to the analysis system. The location of the image acquisition devices is not specifically limited in this embodiment, as long as they can completely capture the teacher's movement trajectory within the preset teaching time period. When determining the teacher movement heatmap within a preset teaching time period based on mobile information, the classroom can first be divided into multiple teaching locations by spatial grids based on the classroom floor plan. Then, the cumulative dwell time and entry frequency of teachers at each teaching location can be statistically analyzed based on the mobile information to obtain the heat intensity value corresponding to each teaching location. Finally, the classroom floor plan can be converted into a color visualization image based on the heat intensity value corresponding to each teaching location. In order to more intuitively display the heat intensity values ​​corresponding to different teaching locations, a gradient color scheme can be used to represent different heat intensity values. For example, green can be used to represent teaching locations with low heat intensity values, yellow can be used to represent teaching locations with medium heat intensity values, and red can be used to represent teaching locations with high heat intensity values. The specific display colors are not limited in this embodiment.

[0030] Step S120: Obtain students' learning status information within the preset teaching time period, and determine the learning status index of each student within the teaching time period based on the learning status information.

[0031] Specifically, learning status information includes, but is not limited to, classroom behavior, classroom interaction, and knowledge mastery. Classroom behavior is mainly reflected in core indicators such as attention span and frequency of distractions. This can be achieved by using image acquisition devices in the classroom to capture images of students, and then using a pre-set AI behavior recognition algorithm to determine each student's head orientation, eye focus, and whether they are leaning on the desk or playing with a pen during a pre-set teaching period. Based on this, the attention span and frequency of distractions for each student during the pre-set teaching period can be determined. Classroom interaction is mainly reflected in core indicators such as the number of times a student speaks and the quality of their questions. This can be achieved by using voice acquisition devices in the classroom to collect each student's speaking data, and then using a pre-set AI voice recognition algorithm to determine the number of times each student actively speaks and the content of their questions during a pre-set teaching period. The quality of questions can be determined based on the relevance between the keywords in the question content and the relevant knowledge points. Knowledge mastery is mainly reflected in core indicators such as answer accuracy and response speed, which can be statistically analyzed based on the accuracy of embedded quizzes and hand-raising responses. After identifying each core indicator, normalize each core indicator to obtain the normalized core indicators.

[0032] After determining the weights for each core indicator based on the course type, a learning status index for each student within a given teaching period is determined based on the normalized core indicators and their corresponding weights. For example, given three core indicators (A, B, and C), the weights for each indicator differ depending on the course type. For a theoretical course, the weights might be A:B:C = 5:3:2; for a practical course, they might be A:B:C = 3:3:4. The course type is relevant to the teaching content delivered by the teacher within the pre-set teaching period.

[0033] Step S130: Determine the teaching coverage map based on the teacher movement heatmap and the learning status index of each student.

[0034] Specifically, the core of determining the teaching coverage map based on teacher movement heatmaps and student learning status indices is to establish a visual correlation between teacher spatial attention and student learning needs. The main steps include heatmap partitioning and weight quantification. The actual student learning situation is aggregated based on the partitioning results, ultimately yielding the teaching coverage map. By linking teacher attention levels with student learning status indices, a connection between teaching and learning is established. Furthermore, by transforming student learning status indices within each heatmap into a visual partition coverage map, abstract teaching-learning correlation data is transformed into intuitive and perceptible visual information. Further, to facilitate teachers' rapid identification of core problem areas and problem types, the specific implementation process of determining the teaching coverage map based on teacher movement heatmaps and each student's learning status index can include: The teacher movement heatmap is partitioned based on the heat intensity values ​​corresponding to each teaching location, resulting in multiple heatmaps, each with a different mean heat intensity. The partition weight for each heatmap is determined based on the mean heat intensity of each heatmap and a pre-defined weight mapping relationship between the mean heat intensity and the partition weight. The partition coverage map for each heatmap is determined based on the learning status index of each student and their corresponding partition weight. Finally, the partition coverage maps of all heatmaps are integrated to obtain the teaching coverage map.

[0035] Specifically, based on a preset feature recognition algorithm, the heat intensity value corresponding to each teaching location is identified from the teacher movement heatmap. After dividing the teacher movement heatmap into regions based on the heat intensity value of each teaching location, multiple heat zones can be obtained. The intensity difference between the heat intensity values ​​corresponding to any two teaching locations in each heat zone is lower than a preset intensity threshold. The specific preset intensity threshold is not specifically limited in this embodiment and can be determined by relevant personnel based on historical experimental data and uploaded to the analysis system. After determining multiple heat zones, the average heat intensity value corresponding to each teaching location within each heat zone is calculated to obtain the average heat intensity value corresponding to each heat zone.

[0036] Different thermal zones correspond to different zone weights. The zone weight for any thermal zone can be determined based on a preset weight mapping relationship. This preset weight mapping relationship is the correspondence between thermal intensity and zone weight. Since the final teaching coverage map is used to reflect teaching quality, and a higher average thermal intensity means that teachers may frequently or for extended periods stay in a particular thermal zone, the students' focus and interaction within that zone may passively increase. Therefore, a higher average thermal intensity corresponds to a lower zone weight. This method facilitates obtaining a more realistic understanding of student behavior. The specific content of the preset weight mapping relationship is not specifically limited in this embodiment and can be determined by relevant personnel based on historical experimental data and uploaded to the analysis system.

[0037] When determining the zone coverage map corresponding to a heat map zone, the mean state index of the learning state index of all students in the heat map zone can be calculated. Then, based on the zone weight corresponding to the heat map zone, the final index corresponding to each heat map zone can be calculated. Different final indices correspond to different coverage colors. After determining the coverage color corresponding to each heat map zone, the coverage color can be integrated into the corresponding heat map zone to obtain the zone coverage map corresponding to each heat map zone. Finally, all the zone coverage maps are stitched together according to the classroom spatial location to obtain the teaching coverage map.

[0038] Furthermore, to improve the accuracy of determining the partition coverage map, before determining the partition coverage map corresponding to a heat map based on the learning status index of each student corresponding to the heat map and the corresponding partition weight, the following may also be included: Identify the learning status indices of any two students within a heatmap zone and determine the difference between their indices. If the difference exceeds a preset threshold, determine the percentage of students with high and low status indices based on the mean status index of the heatmap zone. The mean status index is calculated by averaging the learning status indices of all students within the heatmap zone. Determine the number of students to be removed based on their percentages. Firstly, remove students with high status indices (above the mean) from the high status index group. Secondly, remove students with low status indices (below the mean) from the low status index group. Finally, update the learning status indices of all students within the heatmap zone based on the first and second removal results.

[0039] Specifically, for any partition coverage map, before determining the partition coverage map based on the students' learning status index, the learning status index of all students in the hot partition can be analyzed to determine whether removal processing is required. The index difference between any two students' learning status indices in the hot partition can be calculated first, and it can be determined whether there is an index difference higher than a preset difference threshold. If there is, it indicates that the hot partition needs to be removed. The specific preset difference threshold is not specifically limited in this application embodiment.

[0040] Within each heat zone corresponding to the required elimination process, the mean state index of each student's learning state index is determined. Based on the mean state index and each student's learning state index, the state index type of each student is determined. The state index types include high state index type and low state index type. Specifically, the state index type corresponding to students whose learning state index is higher than the mean state index is high state index type, and the state index type corresponding to students whose learning state index is lower than the mean state index is low state index type. The proportion of students in the high state index type and the low state index type is determined based on the number of students in the high state index type and the low state index type, respectively. Then, the number of students to be eliminated based on the proportion of students in the high state index type and the low state index type is determined. For example, if the proportion of students in the high state index type is 4:1, then the number of students to be eliminated in the high state index type is 4 and the number of students to be eliminated in the low state index type is 1. Alternatively, the number of students to be eliminated in the high state index type is 8 and the number of students to be eliminated in the low state index type is 2. The specific number of students to be eliminated is not specifically limited in this embodiment of the application, as long as the ratio of the number of students to be eliminated in the high state index type to the low state index type is consistent with the proportion of students in the high state index type.

[0041] The first elimination process, based on the number of students with high learning status indices (above the mean), removes students from the heatmap partitions. The second elimination process, based on the number of students with low learning status indices (below the mean), removes students from the heatmap partitions. These first and second elimination results update the learning status indices of each student within the heatmap partition. Finally, the mean learning status index used in determining the partition coverage map is updated based on these elimination results. By detecting differences in student learning status indices, classifying them, removing anomalies, and updating data, precise cleaning and optimization of student status indices are achieved, avoiding significant discrepancies between the determined partition coverage map and actual teaching effectiveness caused by excessive dispersion in student status indices.

[0042] Step S140: Based on the teaching coverage map, determine the areas and teaching content to be focused on, and generate teaching feedback suggestions based on the areas and teaching content to be focused on. The teaching feedback suggestions are used to remind teachers to make teaching adjustments in the next teaching period.

[0043] Specifically, pixel recognition can be used to identify areas with pixel values ​​below a preset pixel threshold from the teaching coverage map corresponding to a preset teaching time period. Based on this, areas of interest can be determined. The specific pixel threshold is not specifically limited in this embodiment and can be determined by relevant personnel based on historical experimental data and uploaded to the analysis system. The teaching content of interest refers to knowledge points, teaching segments, or explanations within the preset teaching time period that are highly correlated with the decline in student performance within the areas of interest where pixel values ​​are below the preset threshold. By extracting time-series data of student learning status indices within the areas of interest, key time points where student status indices decline significantly can be identified. These key time points are then matched with the course schedule, teaching logs, or classroom video clips for the preset teaching time period to pinpoint the core knowledge points, teaching segments, or specific explanations being delivered by the teacher at that key time point. After determining teaching feedback suggestions, these suggestions can be sent to relevant teachers through a prompt window, allowing them to adjust the teaching methods or content for the next teaching period based on the actual teaching effect of the current teaching period.

[0044] In this embodiment, by analyzing the teacher's movement information during the teaching phase and generating a teacher movement heatmap, it is easy to intuitively present the areas of high teacher attention and areas of low attention. This helps teachers objectively recognize the limitations of their teaching behavior, proactively adjust their movement range, and ensure that every student receives basic attention. In addition, by analyzing students' learning status information during the teaching process and quantifying each student's classroom performance in the form of a learning status index, it is easy to overcome the bottleneck of traditional classroom observation, which makes it difficult to pay attention to the details of each student. By combining teacher movement and student learning status, it is easy to break the disconnect between teacher behavior and student feedback. Finally, after determining the areas and students that need attention through the teaching coverage map, targeted teaching adjustment methods are generated, which facilitates the formation of a data-driven teaching optimization closed loop. By directly transforming the analysis results of AI teaching technology into executable teaching actions, it is easy to improve the integration of AI teaching technology into the actual teaching process, thereby improving teaching quality.

[0045] Furthermore, to facilitate teachers in monitoring student performance fluctuations by viewing classroom teaching images, the method provided in this application embodiment also includes: The system acquires real-time learning status information for each student, identifies the status characteristics of each student from each real-time learning status information, and identifies the real-time status characteristic value of each status characteristic; determines the real-time status index of each student based on the preset status characteristic weights and each real-time status characteristic value; generates a learning status feedback bar for each student based on the real-time status index of each student, and overlays each learning status feedback bar onto the corresponding student position in the classroom teaching image to obtain a real-time classroom status feedback image.

[0046] Specifically, the method for obtaining students' real-time status information and determining each student's real-time status index based on this information can refer to the method for determining the learning status index of students within a heat map zone as described in the above embodiment, and will not be repeated here. The status characteristics are similar to the core indicators, including focus duration, frequency of distracting behaviors, number of times spoken, question quality, answer accuracy, and response speed. After determining the real-time status characteristic value of each characteristic, normalization processing is also required.

[0047] When generating a learning status feedback bar for each student based on their real-time status index, a horizontal progress bar design can be adopted. The length of the learning status feedback bar is positively correlated with the student's real-time status index, and the display color can be matched according to the index range to enhance visual perception. For example, green can be displayed when the learning status index is ≥80; yellow can be displayed when the learning status index is between 60 and 79; and red can be displayed when the learning status index is below 60. The specific display color is not specifically limited in this embodiment.

[0048] A preset feature recognition algorithm can be used to identify the position of each student in a classroom teaching image, and the generated learning status feedback bar can be superimposed below or to the right of the corresponding student's position to obtain a real-time classroom status feedback image. The superposition position of the learning status feedback bar is not specifically limited in this embodiment, as long as it does not obscure the student's face. By capturing students' learning status in real time, quantifying the status index, and superimposing it on the classroom teaching image in the form of a feedback bar, it is convenient for relevant teachers to simultaneously grasp the fluctuations in students' status by viewing the classroom teaching image.

[0049] Furthermore, to reduce the time cost for teachers to conduct post-lesson review, the technical solution provided in this application also includes: The system identifies the features of each learning status feedback bar in the real-time classroom status feedback image, including the growth rate and length of the feedback bar. Learning status feedback bars that meet preset anomaly judgment criteria are identified as abnormal feedback bars and marked with stars. The preset anomaly judgment criteria are: the growth rate of the feedback bar is lower than a preset growth rate threshold; or, the length of the feedback bar is lower than a preset length threshold. The system identifies the starred student positions corresponding to each star within a preset observation period. When the distribution density of starred student positions is greater than a preset density threshold and the number of starred students is higher than a preset number threshold, a one-click slicing mode is activated to slice the classroom teaching segment corresponding to the preset observation period, obtaining starred sliced ​​teaching segments. The starred sliced ​​classroom teaching segments are then fed back as feedback.

[0050] Specifically, a preset feature recognition algorithm can be used to detect each learning status feedback bar in the real-time classroom status feedback image, and to identify the growth rate and length of each feedback bar in real time. Feedback bars with a growth rate lower than a preset growth rate threshold, or a length lower than a preset length threshold, are identified as abnormal feedback bars. The specific preset growth rate threshold and preset length threshold are not specifically limited in this embodiment and can be determined by relevant personnel based on historical experimental data and uploaded to the analysis system. After identifying abnormal feedback bars, a star mark can be superimposed on the upper right or upper left corner of the abnormal feedback bar, and the star timestamp and the corresponding student's position can be recorded.

[0051] The preset observation period is a period of time after the first abnormal feedback bar is identified. The duration of the preset observation period is not specifically limited in this embodiment. The distribution density of all starred student positions is statistically analyzed to determine whether there is a group abnormality. If the distribution density of starred student positions is greater than the preset density threshold and the number of starred students is higher than the preset number threshold, it indicates that there is a group abnormality. At this time, the one-click slicing mode can be activated to slice the classroom teaching segments corresponding to the preset observation period to obtain star slice teaching segments. Based on the preset observation period, the corresponding classroom teaching segments can be extracted from complete classroom recordings and identified as star slice teaching segments. By automatically editing the classroom teaching segments of the preset observation period, the core segments containing the "problem scenario" can be quickly obtained without the need for teachers to manually select them after class, thereby reducing the time cost for teachers to conduct teaching reviews after class.

[0052] Furthermore, the technical method provided in this application embodiment also includes steps S210-S240, such as... Figure 2 As shown, where: Step S210: Obtain the initial classroom teaching records and identify the teacher behavior occupancy rate and teacher-student interaction frequency corresponding to each teaching period in the initial classroom teaching records.

[0053] Specifically, the initial classroom teaching records contain continuous full-length video and audio, which may include a large amount of non-core content, such as disciplinary emphasis and transitional segments. Directly using these initial records for post-class review could require significant time to filter out the core content. Therefore, the initial classroom teaching records need to be processed based on the teacher behavior occupancy rate and teacher-student interaction frequency for each teaching segment. The teacher behavior occupancy rate refers to the proportion of core teacher behaviors (such as lecturing, questioning, and demonstrating) at a given teaching moment, used to determine whether that moment is a teacher-led core teaching segment. The teacher-student interaction frequency refers to the number of effective interactions between teachers and students during a given teaching segment, used to determine whether that moment is a core interactive segment with high classroom participation. This can be determined by performing audio and video dual-modal recognition on the initial classroom teaching records; the specific determination method is not specifically limited in this embodiment. The duration of each teaching segment can be 8 minutes, 12 minutes, etc., and the specific duration is not specifically limited in this embodiment.

[0054] Step S220: Based on the teacher behavior occupancy rate and teacher-student interaction frequency corresponding to each teaching period, determine the time period classroom type and time period teaching content corresponding to each teaching period, and determine the slice display format type of each teaching period according to the time period classroom type corresponding to each teaching period.

[0055] Specifically, different teacher behavior rates and teacher-student interaction frequencies correspond to different time-segment classroom types, which can be determined based on a pre-set Rt-Ch diagram. Time-segment classroom types include dialogue-based, practice-based, blended, and lecture-based types. The teaching content for each time segment is the time-segment teaching record corresponding to the initial classroom teaching record. The slice display format type corresponding to each teaching time segment's classroom type can be determined based on a pre-set format type mapping relationship. Slice display format types can include, but are not limited to, audio, dialogue text, video, and whiteboard writing. The pre-set format type mapping relationship is the correspondence between time-segment classroom types and slice display format types, which can be determined by relevant staff based on historical experimental data and uploaded to the analysis system in advance.

[0056] Step S230: Based on the teaching content corresponding to each teaching period, determine the teaching focus markers and comprehensive learning status index for each teaching period. Based on the teaching focus markers and comprehensive learning status index, determine the slice display level type corresponding to each teaching period. The comprehensive learning status index is determined by the learning status index of each student in the corresponding teaching period.

[0057] Specifically, based on a preset feature recognition algorithm, key teaching indicators can be identified and determined from the teaching content corresponding to the teaching period. These indicators can include key knowledge points, error-prone knowledge points, and key chapters. Different key indicators correspond to different types of key points, including high-key, secondary-key, and non-key. A preset key mapping relationship can be used to determine the key type corresponding to any key indicator. This preset mapping relationship is the correspondence between key indicators and key types. The method for determining the learning status index based on the teaching content of the teaching period is not elaborated here; refer to the method for determining the student's learning status index during the teaching period in the above embodiment. The comprehensive learning status index is the sum of the average learning status indices of all students during the teaching period. Different key teaching indicators and comprehensive learning status indices correspond to different slice display levels, including detailed display, overview display, and thumbnail display, for example: The slice display level corresponding to the high-emphasis and low-emphasis periods is detailed display, that is, the explanation process, student reactions and teacher interaction of the corresponding teaching period are fully preserved to facilitate the review of problems. Among them, the low-emphasis period is the teaching period when the comprehensive learning state index is lower than the preset state index threshold. The slice display level type corresponding to the secondary focus and high state period is overview display, that is, only the core conclusions and timestamps are retained, and redundant information is simplified. Among them, the high state period is the teaching period when the comprehensive learning state index is higher than the preset state index threshold. The slice display level for non-key and normal periods is a thumbnail display, that is, text summaries and keyframe screenshots can be used to replace the complete teaching record. Among them, the normal period is the teaching period when the comprehensive learning status index is equal to the preset status index threshold.

[0058] Step S240: Based on the slice display format type, slice display degree type and teaching content of each teaching period, determine the teaching record of each teaching period slice, and integrate the teaching records of all teaching periods slice to obtain the target classroom teaching record corresponding to the initial classroom teaching record.

[0059] Specifically, based on the display format and level of each teaching segment, the teaching content for each segment is edited accordingly, generating segmented teaching records for each teaching segment. This ensures that the format matches the content and the level matches the value. For example, in lecture-style teaching segments, the display format includes explanation footage and blackboard writing, and the level of detail is shown. The corresponding segmented teaching record is a continuous video of the teacher's explanation, with synchronized speech-to-text subtitles. Additionally, key frames from the blackboard or PPT are inserted via split-screen. Finally, all initial classroom teaching records are integrated according to the teaching segments to obtain the target classroom teaching record.

[0060] In this embodiment of the application, by analyzing the teacher behavior occupancy rate and teacher-student interaction frequency corresponding to each teaching moment, it is convenient to divide the entire teaching process into time periods. Based on the actual teaching situation corresponding to each teaching period, the slice display format type and slice display degree type are determined for each teaching period. Finally, the teaching records of each teaching period slice are integrated. This method of dividing records helps to avoid the redundancy problem of full recording. Through AI teaching technology, classroom records are transformed from passive storage into actively optimized structured resources, making the records more in line with teaching needs, thereby facilitating the efficient accumulation of teaching resources and the continuous improvement of teaching quality.

[0061] This application provides an analysis system, such as... Figure 3 As shown, Figure 3 The analysis system 300 shown includes a processor 301 and a memory 303. The processor 301 and the memory 303 are connected, for example, via a bus 302. Optionally, the analysis system 300 may also include a transceiver 304. It should be noted that in practical applications, the transceiver 304 is not limited to one type, and the structure of this analysis system 300 does not constitute a limitation on the embodiments of this application.

[0062] Processor 301 may be a CPU (Central Processing Unit), a general-purpose processor, a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. It can implement or execute the various exemplary logic blocks, modules, and circuits described in conjunction with the disclosure of this application. Processor 301 may also be a combination that implements computational functions, such as including one or more microprocessor combinations, a combination of a DSP and a microprocessor, etc.

[0063] Bus 302 may include a pathway for transmitting information between the aforementioned components. Bus 302 may be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus, etc. Bus 302 can be divided into address bus, data bus, control bus, etc. For ease of representation, Figure 3 The symbol is represented by only one line, but this does not mean that there is only one bus or one type of bus.

[0064] The memory 303 may be a ROM (Read Only Memory) or other type of static storage device capable of storing static information and instructions, RAM (Random Access Memory) or other type of dynamic storage device capable of storing information and instructions, or an EEPROM (Electrically Erasable Programmable Read Only Memory), CD-ROM (Compact Disc Read Only Memory) or other optical disc storage, optical disc storage (including compressed optical discs, laser discs, optical discs, digital universal optical discs, Blu-ray discs, etc.), magnetic disk storage media or other magnetic storage devices, or any other medium capable of carrying or storing desired program code in the form of instructions or data structures and accessible by a computer, but not limited thereto.

[0065] The memory 303 is used to store application code that executes the solution of this application, and its execution is controlled by the processor 301. The processor 301 is used to execute the application code stored in the memory 303 to implement the content shown in the foregoing method embodiments.

[0066] The analysis system includes, but is not limited to, mobile terminals such as mobile phones, laptops, digital radio receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), and in-vehicle terminals (such as in-vehicle navigation terminals), as well as fixed terminals such as digital TVs and desktop computers. It can also include servers. Figure 3 The analysis system shown is merely an example and should not impose any limitations on the functionality and scope of use of the embodiments of this application.

[0067] This application provides a computer-readable storage medium storing a computer program that, when run on a computer, enables the computer to execute the corresponding content in the aforementioned method embodiments.

[0068] This application provides a computer program product including a computer program that, when executed by a processor, implements the methods described in any of the above embodiments.

[0069] It should be understood that although the steps in the flowcharts of the accompanying figures are shown sequentially as indicated by the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the accompanying figures may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times, and their execution order is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the sub-steps or stages of other steps.

[0070] The above description is only a partial embodiment of this application. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of this application, and these improvements and modifications should also be considered within the scope of protection of this application.

Claims

1. A method of teaching data analysis, characterized by, The method comprises the following steps: obtaining the movement information of a teacher in a preset teaching time period, and determining a teacher walking heat map of the teacher in the preset teaching time period based on the movement information; obtaining the learning state information of students in the preset teaching time period, and determining the learning state index of each student in the teaching time period based on the learning state information; determining a teaching coverage map based on the teacher walking heat map and the learning state index of each student; determining a region to be concerned and teaching content to be concerned based on the teaching coverage map, and generating a teaching feedback suggestion based on the region to be concerned and the teaching content to be concerned, the teaching feedback suggestion being used to remind the teacher to make teaching adjustment in the next teaching time period.

2. The method of claim 1, wherein, The method of determining the teaching coverage map based on the teacher walking heat map and the learning state index of each student comprises the following steps: performing zoning processing on the teacher walking heat map based on the heat intensity values corresponding to the respective teaching positions, to obtain a plurality of heat zones, each heat zone corresponding to a different heat intensity mean value; determining the zone weight corresponding to each heat zone based on the heat intensity mean value of each heat zone and a preset weight mapping relationship, the preset weight mapping relationship being a corresponding relationship between the heat intensity mean value and the zone weight; determining the zone coverage map corresponding to each heat zone based on the learning state index of each student corresponding to each heat zone and the corresponding zone weight; integrating the zone coverage maps of all heat zones to obtain the teaching coverage map.

3. The method of claim 2, wherein, Before determining the zone coverage map corresponding to the heat zone based on the learning state index of each student corresponding to the heat zone and the corresponding zone weight, the method further comprises the following steps: identifying the learning state indexes of any two students in the heat zone, and determining the index difference value between the learning state indexes of the two students; if the index difference value between the learning state indexes of any two students is higher than a preset difference threshold, determining the student population proportion of the high state index type and the low state index type based on the mean state index corresponding to the heat zone, the mean state index being determined by mean calculation of the learning state indexes of the students in the heat zone; determining the high state elimination population and the low state elimination population based on the student population proportion; performing first elimination from the students of the high state index type whose learning state index is higher than the mean state based on the high state elimination population, to obtain a first elimination result; performing second elimination from the students of the low state index type whose learning state index is lower than the mean state based on the low state elimination population, to obtain a second elimination result; updating the learning state index of each student corresponding to the heat zone based on the first elimination result and the second elimination result.

4. The method of claim 1, wherein, The method further comprises the following steps: obtaining the real-time learning state information of each student, identifying the state feature of each student from each real-time learning state information, and identifying the real-time state feature value of each state feature; determining the real-time state index of each student based on a preset state feature weight and the respective real-time state feature values. A learning state feedback bar corresponding to each student is generated based on the real-time state index of each student, and each learning state feedback bar is superimposed on the position of the corresponding student in the classroom teaching image to obtain a real-time classroom state feedback image.

5. The method of claim 4, wherein, Further comprising: identifying a feedback bar feature of each learning state feedback bar in the real-time classroom state feedback image, the feedback bar feature including a feedback bar growth rate and a feedback bar length; determining a learning state feedback bar that meets a preset abnormality judgment standard as an abnormal feedback bar, and superimposing a star mark on the abnormal feedback bar, the preset abnormality judgment standard being that the feedback bar growth rate is lower than a preset growth rate threshold, or the feedback bar length is lower than a preset length threshold; identifying a star mark student position corresponding to each star mark in a preset observation time period, and starting a one-key slicing mode when the distribution density of the star mark student position is greater than a preset density threshold and the number of star mark students is higher than a preset number threshold, the one-key slicing mode being used to slice a classroom teaching segment corresponding to the preset observation time period to obtain a star mark sliced teaching segment; feeding back the star mark sliced classroom teaching segment.

6. The method of claim 1, wherein, Further comprising: obtaining an initial classroom teaching record and identifying a teacher behavior occupancy rate and a teacher-student interaction frequency corresponding to each teaching period in the initial classroom teaching record; determining a period classroom type and a period teaching content corresponding to each teaching period based on the teacher behavior occupancy rate and the teacher-student interaction frequency corresponding to each teaching period, and determining a slicing display form type of each teaching period according to the period classroom type corresponding to each teaching period; determining a teaching focus identifier and a comprehensive learning state index of each teaching period based on the period teaching content corresponding to each teaching period, and determining a slicing display degree type corresponding to each teaching period based on the teaching focus identifier and the comprehensive learning state index, the comprehensive learning state index being determined by the learning state index of each student in the corresponding teaching period; determining a period sliced teaching record corresponding to each teaching period based on the slicing display form type, the slicing display degree type, and the period teaching content corresponding to each teaching period, and integrating the period sliced teaching records corresponding to all teaching periods to obtain a target classroom teaching record corresponding to the initial classroom teaching record.

7. An analytical system, characterized by The analysis system comprises: at least one processor; a memory; at least one application program, wherein the at least one application program is stored in the memory and is configured to be executed by the at least one processor, and the at least one application program is configured to execute a teaching data analysis method according to any one of claims 1-6.

8. A computer-readable storage medium, characterized in that, comprising: a computer program stored in the memory and capable of being loaded and executed by the processor to implement a teaching data analysis method according to any one of claims 1-6.

9. A computer program product, characterised in that, comprising a computer program, which, when executed by a processor, implements the steps of a teaching data analysis method according to any one of claims 1-6.

Images