A public health teaching assessment and feedback system

Abstract

The present application relates to teaching quality evaluation, in particular to a public health teaching evaluation and feedback system, a control unit collects teaching scene data through a teaching scene data collection module and carries out pretreatment to obtain a multi-modal data stream, the control unit extracts the spatio-temporal features of video data and the feature vector of audio data through a first feature extraction module, and utilizes a target feature acquisition module to carry out dimension reduction processing on the spatio-temporal features and the feature vector to obtain target features, the control unit extracts teacher behavior features, student behavior features and interactive behavior features from the target features through a second feature extraction module, and utilizes an index parameter acquisition module to determine index parameters according to the teacher behavior features, the student behavior features and the interactive behavior features, and the control unit obtains a first teaching quality evaluation result according to the index parameters through a first teaching quality evaluation module; the present application can overcome the defect that it is difficult to comprehensively, accurately and efficiently evaluate the teaching quality of public health teaching.

Description

Technical Field

[0001] This invention relates to teaching quality assessment, specifically to a public health teaching assessment and feedback system. Background Technology

[0002] Public health education combines theoretical knowledge transmission with practical skills development, making teaching quality assessment a crucial step in controlling talent cultivation levels and optimizing teaching methods. Currently, the teaching quality assessment methods used in the field have many shortcomings and are ill-suited to the evolving needs of public health education. Traditional teaching quality assessments are largely based on written exam scores, class attendance, and manual lecture evaluations, resulting in a singular evaluation dimension, an overemphasis on final teaching outcomes, and a neglect of dynamic details such as teacher-student behavior and interactive communication in the classroom. Subjective judgment plays a significant role, leading to insufficient objectivity and reference value in the assessments.

[0003] Meanwhile, existing teaching quality assessments struggle to efficiently integrate multimodal teaching data such as classroom videos and audio recordings. Conventional analysis methods cannot accurately extract teacher teaching behaviors and student classroom response characteristics. The dynamic interaction relationships and intensity between teachers and students cannot be quantitatively represented, and the evolutionary patterns of time-series behaviors cannot be effectively explored.

[0004] Furthermore, current teaching quality assessment systems largely separate classroom performance from after-class learning outcomes, failing to link classroom performance data with academic achievement data for analysis, thus hindering the formation of a comprehensive and unified evaluation basis. The assessment and analysis efficiency is low, unable to output assessment results in real time, unable to promptly identify problems in the teaching process, and unable to adjust teaching strategies in a targeted manner, thus restricting the steady improvement of public health teaching quality. Therefore, there is an urgent need to build an intelligent, multi-dimensional public health teaching assessment and feedback system. Summary of the Invention

[0005] (a) Technical problems to be solved

[0006] In view of the above-mentioned shortcomings of the existing technology, the present invention provides a public health teaching evaluation and feedback system, which can effectively overcome the defects of the existing technology in that it is difficult to conduct a comprehensive, accurate and efficient evaluation of the teaching quality of public health teaching.

[0007] (II) Technical Solution

[0008] To achieve the above objectives, the present invention provides the following technical solution:

[0009] A public health teaching evaluation and feedback system includes a control unit. The control unit collects teaching scenario data through a teaching scenario data acquisition module and preprocesses it to obtain a multimodal data stream. The control unit extracts spatiotemporal features of video data and feature vectors of audio data through a first feature extraction module, and performs dimensionality reduction processing on the spatiotemporal features and feature vectors using a target feature acquisition module to obtain target features. The control unit extracts teacher behavior features, student behavior features, and interactive behavior features from the target features through a second feature extraction module, and determines indicator parameters based on the teacher behavior features, student behavior features, and interactive behavior features using an indicator parameter acquisition module. The control unit performs teaching quality evaluation based on the indicator parameters through a first teaching quality evaluation module to obtain a first teaching quality evaluation result.

[0010] The control unit collects teaching outcome data through the data acquisition module and preprocesses it to obtain a teaching outcome dataset. The control unit performs in-depth analysis of the teaching outcome dataset through the data analysis module to obtain teaching evaluation data. It then uses the second teaching quality evaluation module to conduct a teaching quality evaluation based on the teaching evaluation data to obtain a second teaching quality evaluation result. Finally, the control unit outputs the teaching evaluation result by combining the first and second teaching quality evaluation results through the real-time feedback module.

[0011] Preferably, the teaching scenario data acquisition module acquires teaching scenario data and preprocesses it to obtain a multimodal data stream, including:

[0012] Use an audio and video data acquisition system to collect raw video and audio data of teaching scenarios;

[0013] The original video data is subjected to noise reduction and image enhancement processing to obtain the target video data;

[0014] The original audio data is subjected to noise reduction, sound source localization, and speech enhancement to obtain the target audio data.

[0015] Align the target video data and target audio data based on timestamps to obtain a multimodal data stream.

[0016] Preferably, the first feature extraction module extracts spatiotemporal features of the video data, including:

[0017] The target video data is segmented into consecutive frames to obtain multiple sub-video information segments;

[0018] The information from each video segment is input into the first feature extraction model to obtain the spatiotemporal features of each video segment.

[0019] Preferably, the first feature extraction module extracts feature vectors from the audio data, including:

[0020] Short-time Fourier transform is performed on the target audio data corresponding to each video segment to obtain the multi-segment audio spectrogram;

[0021] A convolutional network is used to extract features from the spectrograms of each audio segment, resulting in feature vectors for each segment.

[0022] Preferably, the second feature extraction module extracts teacher behavior features from the target features, including:

[0023] The target features are input into the second feature extraction model, which uses a sliding window to perform continuous frame analysis to obtain the initial behavioral features of the teacher at different time scales.

[0024] The spatial attention module in the second feature extraction model is used to determine the attention weights for different spatial regions;

[0025] Teacher behavioral characteristics are identified based on initial behavioral features and attention weights.

[0026] Preferably, the second feature extraction module extracts interactive behavior features from the target features, including:

[0027] Determine the interaction relationship between teachers and students based on the characteristics of teacher behavior and student behavior;

[0028] A dynamic interaction graph is constructed with teachers and students as nodes. The edges of the dynamic interaction graph represent interaction relationships, and the weight of each edge is matched with the interaction intensity of the corresponding interaction relationship.

[0029] The dynamic interaction graph is input into the spatiotemporal graph convolutional network, which extracts the spatiotemporal features of the interaction relationship to obtain the interaction feature vector of each node.

[0030] We use a Long Short-Term Memory (LSTM) neural network to analyze the interaction feature vectors of each node and extract the interaction behavior features.

[0031] Preferably, the indicator parameter acquisition module determines the indicator parameters based on teacher behavior characteristics, student behavior characteristics, and interaction behavior characteristics, including:

[0032] Based on teacher behavior characteristics, student behavior characteristics, and interaction behavior characteristics, determine the distribution of student emotional state, classroom noise level, student attention span, student enthusiasm for answering questions, student participation behavior, teacher questioning frequency, student response rate, and duration of teacher-student interaction.

[0033] The classroom atmosphere index is determined based on the distribution of students' emotional states, classroom noise levels, preset weights for emotional states, and preset weights for noise levels.

[0034] The cumulative calculation method of time decay was used to analyze students' attention duration, students' enthusiasm for answering questions and students' participation behavior, and to determine the student participation index.

[0035] The frequency of teacher questions, student response rate, and duration of teacher-student interaction were analyzed to determine indicators of teaching interaction effectiveness.

[0036] Preferably, the data acquisition module collects teaching outcome data and performs preprocessing to obtain a teaching outcome dataset, including:

[0037] Collect teaching outcome data including student attendance records, learning duration, frequency of classroom interaction, homework submission rate, homework accuracy rate, and test scores;

[0038] The teaching outcome data is processed by denoising, missing value completion, outlier removal, and format quantization to obtain the teaching outcome dataset.

[0039] Preferably, the data analysis module performs in-depth analysis on the teaching outcome dataset to obtain teaching evaluation data, including:

[0040] By integrating student attendance records, learning duration, and classroom interaction frequency, the single learning duration corresponding to each attendance and the classroom interaction frequency corresponding to each attendance are obtained. The single learning duration is corrected based on the classroom interaction frequency, and the average of the corrected single learning durations of all students is obtained to get the effective teaching duration.

[0041] The homework submission rate, homework accuracy, and test scores are quantified to obtain homework completion rate and test accuracy. The homework completion rate and test accuracy are weighted to obtain knowledge mastery rate. The knowledge mastery rate of all students is estimated using a tracking model to obtain the teaching mastery rate.

[0042] Preferably, the second teaching quality assessment module performs a teaching quality assessment based on the teaching assessment data to obtain a second teaching quality assessment result, including:

[0043] Compare the effective teaching time with the total pre-set course time to calculate the class completion rate index;

[0044] Compare the teaching mastery with the pre-set knowledge mastery to calculate the knowledge attainment rate index;

[0045] Based on the preset teaching objective weighting table, the class completion rate and knowledge attainment rate are weighted and integrated to calculate the comprehensive teaching effectiveness score.

[0046] The comprehensive teaching effectiveness score is matched with the preset evaluation level range to obtain the second teaching quality assessment result.

[0047] (III) Beneficial Effects

[0048] Compared with existing technologies, the public health teaching assessment and feedback system provided by this invention has the following beneficial effects:

[0049] 1) Integrate multi-source data to improve the comprehensiveness of the assessment.

[0050] The system synchronously collects data on classroom teaching scenarios and post-class teaching outcomes, breaking the limitations of traditional assessments that only focus on a single dimension. On the one hand, it relies on multimodal data such as video and audio to deeply explore process details such as classroom atmosphere, student participation, and teaching interaction; on the other hand, it integrates outcome data such as attendance, homework, and tests to calculate effective teaching time and teaching mastery. By combining real-time classroom performance with final learning outcomes, it constructs a dual assessment system, avoids subjective bias in human evaluation, and comprehensively covers the teaching process, making the basis for judging teaching quality more complete and detailed.

[0051] 2) Intelligent algorithm analysis enhances evaluation accuracy.

[0052] The system is equipped with multiple deep learning models to sequentially complete feature extraction, behavior recognition, and interaction relationship modeling. With the help of these models, it automatically splits frame sequences, extracts spatiotemporal features from video data and feature vectors from audio data, quantifies the interaction intensity of interactive relationships through dynamic interaction graphs, and analyzes behavioral change patterns by combining temporal networks. At the same time, it standardizes the processing of various teaching outcome data, automatically calculates multiple indicators such as class completion rate and knowledge attainment rate, effectively replacing inefficient manual statistical analysis, accurately capturing subtle differences in teaching behavior, objectively quantifying various indicator parameters, and significantly improving the accuracy of teaching quality assessment.

[0053] 3) Linking dual results enhances the practical value of feedback.

[0054] The system generates assessment results at both the classroom behavior and teaching effectiveness levels. By merging and summarizing these two types of results, a final teaching assessment is output. This provides a clear picture of shortcomings in the teaching process, weaknesses in students' classroom learning, and actual knowledge absorption. It helps instructors quickly identify teaching problems, adjust the pace of instruction, interaction methods, and key teaching points in a timely manner, and abandon outdated and one-sided evaluation models. This enables real-time output of teaching assessments, providing a reliable basis for optimizing public health teaching and adjusting teaching strategies, and effectively contributing to the steady improvement of the overall quality of public health teaching. Attached Figure Description

[0055] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are merely some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without any creative effort.

[0056] Figure 1 This is a schematic diagram of the system of the present invention;

[0057] Figure 2 This is a schematic diagram of the process for obtaining the second teaching quality assessment result in this invention. Detailed Implementation

[0058] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0059] The following describes the specific functional modules of the public health teaching assessment and feedback system provided by this invention, using concrete examples (such as...). Figure 1 (as shown) and technical effects.

[0060] The system's functional modules include: a control unit, which collects teaching scenario data through a teaching scenario data acquisition module and preprocesses it to obtain a multimodal data stream; the control unit extracts spatiotemporal features of video data and feature vectors of audio data through a first feature extraction module, and uses a target feature acquisition module to perform dimensionality reduction processing on the spatiotemporal features and feature vectors to obtain target features; the control unit extracts teacher behavior features, student behavior features, and interactive behavior features from the target features through a second feature extraction module, and uses an indicator parameter acquisition module to determine indicator parameters based on the teacher behavior features, student behavior features, and interactive behavior features; and the control unit performs teaching quality evaluation based on the indicator parameters through a first teaching quality evaluation module to obtain the first teaching quality evaluation result.

[0061] I. Teaching Scenario Data Acquisition Module

[0062] The teaching scenario data acquisition module collects teaching scenario data and preprocesses it to obtain a multimodal data stream, including:

[0063] Use an audio and video data acquisition system to collect raw video and audio data of teaching scenarios;

[0064] The original video data is subjected to noise reduction and image enhancement processing to obtain the target video data;

[0065] The original audio data is subjected to noise reduction, sound source localization, and speech enhancement to obtain the target audio data.

[0066] Align the target video data and target audio data based on timestamps to obtain a multimodal data stream.

[0067] II. First Feature Extraction Module

[0068] The first feature extraction module extracts the spatiotemporal features of the video data, including:

[0069] The target video data is segmented into consecutive frames to obtain multiple sub-video information segments;

[0070] The information from each video segment is input into the first feature extraction model to obtain the spatiotemporal features of each video segment.

[0071] The first feature extraction module extracts feature vectors from the audio data, including:

[0072] Short-time Fourier transform is performed on the target audio data corresponding to each video segment to obtain the multi-segment audio spectrogram;

[0073] A convolutional network is used to extract features from the spectrograms of each audio segment, resulting in feature vectors for each segment.

[0074] III. Second Feature Extraction Module

[0075] The second feature extraction module extracts teacher behavior features from the target features, including:

[0076] The target features are input into the second feature extraction model, which uses a sliding window to perform continuous frame analysis to obtain the initial behavioral features of the teacher at different time scales.

[0077] The spatial attention module in the second feature extraction model is used to determine the attention weights for different spatial regions;

[0078] Teacher behavioral characteristics are identified based on initial behavioral features and attention weights.

[0079] The second feature extraction module extracts interactive behavior features from the target features, including:

[0080] Determine the interaction relationship between teachers and students based on the characteristics of teacher behavior and student behavior;

[0081] A dynamic interaction graph is constructed with teachers and students as nodes. The edges of the dynamic interaction graph represent interaction relationships, and the weight of each edge is matched with the interaction intensity of the corresponding interaction relationship.

[0082] The dynamic interaction graph is input into the spatiotemporal graph convolutional network, which extracts the spatiotemporal features of the interaction relationship to obtain the interaction feature vector of each node.

[0083] We use a Long Short-Term Memory (LSTM) neural network to analyze the interaction feature vectors of each node and extract the interaction behavior features.

[0084] IV. Indicator Parameter Acquisition Module

[0085] The indicator parameter acquisition module determines indicator parameters based on teacher behavior characteristics, student behavior characteristics, and interaction behavior characteristics, including:

[0086] Based on teacher behavior characteristics, student behavior characteristics, and interaction behavior characteristics, determine the distribution of student emotional state, classroom noise level, student attention span, student enthusiasm for answering questions, student participation behavior, teacher questioning frequency, student response rate, and duration of teacher-student interaction.

[0087] The classroom atmosphere index is determined based on the distribution of students' emotional states, classroom noise levels, preset weights for emotional states, and preset weights for noise levels.

[0088] The cumulative calculation method of time decay was used to analyze students' attention duration, students' enthusiasm for answering questions and students' participation behavior, and to determine the student participation index.

[0089] The frequency of teacher questions, student response rate, and duration of teacher-student interaction were analyzed to determine indicators of teaching interaction effectiveness.

[0090] The above technical solution involves a system equipped with multiple deep learning models that sequentially complete feature extraction, behavior recognition, and interaction relationship modeling. The system automatically splits frame sequences, extracts spatiotemporal features from video data and feature vectors from audio data, quantifies the interaction intensity of interactive relationships through dynamic interaction graphs, and analyzes behavioral change patterns by combining temporal networks.

[0091] The system's functional modules also include: the control unit collects teaching outcome data through the data acquisition module and preprocesses it to obtain a teaching outcome dataset; the control unit performs in-depth analysis of the teaching outcome dataset through the data analysis module to obtain teaching evaluation data; and uses the second teaching quality evaluation module to conduct teaching quality evaluation based on the teaching evaluation data to obtain the second teaching quality evaluation result; and the control unit outputs the teaching evaluation result by combining the first and second teaching quality evaluation results through the real-time feedback module.

[0092] V. Data Acquisition Module

[0093] The data acquisition module collects teaching outcome data and preprocesses it to obtain a teaching outcome dataset, such as... Figure 2 As shown, it includes:

[0094] Collect teaching outcome data including student attendance records, learning duration, frequency of classroom interaction, homework submission rate, homework accuracy rate, and test scores;

[0095] The teaching outcome data is processed by denoising, missing value completion, outlier removal, and format quantization to obtain the teaching outcome dataset.

[0096] VI. Data Analysis Module

[0097] The data analysis module performs in-depth analysis of the teaching outcome dataset to obtain teaching evaluation data, such as... Figure 2 As shown, it includes:

[0098] By integrating student attendance records, learning duration, and classroom interaction frequency, the single learning duration corresponding to each attendance and the classroom interaction frequency corresponding to each attendance are obtained. The single learning duration is corrected based on the classroom interaction frequency, and the average of the corrected single learning durations of all students is obtained to get the effective teaching duration.

[0099] The homework submission rate, homework accuracy, and test scores are quantified to obtain homework completion rate and test accuracy. The homework completion rate and test accuracy are weighted to obtain knowledge mastery rate. The knowledge mastery rate of all students is estimated using a tracking model to obtain the teaching mastery rate.

[0100] VII. Second Teaching Quality Assessment Module

[0101] The second teaching quality assessment module evaluates teaching quality based on the teaching assessment data, and obtains the second teaching quality assessment results, such as... Figure 2 As shown, it includes:

[0102] Compare the effective teaching time with the total pre-set course time to calculate the class completion rate index;

[0103] Compare the teaching mastery with the pre-set knowledge mastery to calculate the knowledge attainment rate index;

[0104] Based on the preset teaching objective weighting table, the class completion rate and knowledge attainment rate are weighted and integrated to calculate the comprehensive teaching effectiveness score.

[0105] The comprehensive teaching effectiveness score is matched with the preset evaluation level range to obtain the second teaching quality assessment result.

[0106] The above-mentioned technical solution standardizes the processing of various teaching outcome data, automatically calculates multiple indicators such as class completion rate and knowledge attainment rate, effectively replaces inefficient manual statistical analysis, can accurately capture subtle differences in teaching behavior, objectively quantify various indicator parameters, and significantly improve the accuracy of teaching quality assessment.

[0107] In this technical solution, the system simultaneously collects classroom teaching scenario data and after-class teaching outcome data, breaking the limitations of traditional assessments that only focus on a single dimension. On the one hand, it relies on multimodal data such as video and audio to deeply explore process details such as classroom atmosphere, student participation, and teaching interaction; on the other hand, it integrates outcome data such as attendance, homework, and tests to calculate effective teaching time and teaching mastery. By combining real-time classroom performance with final learning outcomes, a dual assessment system is constructed to avoid subjective bias in human evaluation, comprehensively cover the teaching process, and make the basis for judging teaching quality more complete and detailed.

[0108] Simultaneously, the system generates assessment results at both the classroom behavior and teaching effectiveness levels. By merging and summarizing these two types of assessment results, the final teaching assessment result is output in a unified manner. This can intuitively reflect the deficiencies in the teaching process, the shortcomings in students' classroom learning status, and the actual situation of knowledge absorption. It helps instructors quickly identify teaching problems, adjust the teaching pace, interaction methods, and teaching focus and difficulties in a timely manner, abandon the outdated and one-sided evaluation model, and achieve real-time output of teaching assessments. This provides a reliable basis for optimizing public health teaching and adjusting teaching strategies, and effectively helps to steadily improve the overall teaching quality of public health.

[0109] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions will not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A public health teaching assessment and feedback system, characterized in that: The system includes a control unit. The control unit collects teaching scenario data through a teaching scenario data acquisition module and preprocesses it to obtain a multimodal data stream. The control unit extracts spatiotemporal features of video data and feature vectors of audio data through a first feature extraction module, and performs dimensionality reduction processing on the spatiotemporal features and feature vectors using a target feature acquisition module to obtain target features. The control unit extracts teacher behavior features, student behavior features, and interactive behavior features from the target features through a second feature extraction module, and determines indicator parameters based on the teacher behavior features, student behavior features, and interactive behavior features using an indicator parameter acquisition module. The control unit performs teaching quality evaluation based on the indicator parameters through a first teaching quality evaluation module to obtain a first teaching quality evaluation result. The control unit collects teaching outcome data through the data acquisition module and preprocesses it to obtain a teaching outcome dataset. The control unit performs in-depth analysis of the teaching outcome dataset through the data analysis module to obtain teaching evaluation data. It then uses the second teaching quality evaluation module to conduct a teaching quality evaluation based on the teaching evaluation data to obtain a second teaching quality evaluation result. Finally, the control unit outputs the teaching evaluation result by combining the first and second teaching quality evaluation results through the real-time feedback module.

2. The public health teaching assessment and feedback system according to claim 1, characterized in that: The teaching scenario data acquisition module collects teaching scenario data and preprocesses it to obtain a multimodal data stream, including: Use an audio and video data acquisition system to collect raw video and audio data of teaching scenarios; The original video data is subjected to noise reduction and image enhancement processing to obtain the target video data; The original audio data is subjected to noise reduction, sound source localization, and speech enhancement to obtain the target audio data. Align the target video data and target audio data based on timestamps to obtain a multimodal data stream.

3. The public health teaching assessment and feedback system according to claim 2, characterized in that: The first feature extraction module extracts spatiotemporal features from the video data, including: The target video data is segmented into consecutive frames to obtain multiple sub-video information segments; The information from each video segment is input into the first feature extraction model to obtain the spatiotemporal features of each video segment.

4. The public health teaching assessment and feedback system according to claim 3, characterized in that: The first feature extraction module extracts feature vectors from the audio data, including: Short-time Fourier transform is performed on the target audio data corresponding to each video segment to obtain the multi-segment audio spectrogram; A convolutional network is used to extract features from the spectrograms of each audio segment, resulting in feature vectors for each segment.

5. The public health teaching assessment and feedback system according to claim 4, characterized in that: The second feature extraction module extracts teacher behavior features from the target features, including: The target features are input into the second feature extraction model, which uses a sliding window to perform continuous frame analysis to obtain the initial behavioral features of the teacher at different time scales. The spatial attention module in the second feature extraction model is used to determine the attention weights for different spatial regions; Teacher behavioral characteristics are identified based on initial behavioral features and attention weights.

6. The public health teaching assessment and feedback system according to claim 5, characterized in that: The second feature extraction module extracts interactive behavior features from the target features, including: Determine the interaction relationship between teachers and students based on the characteristics of teacher behavior and student behavior; A dynamic interaction graph is constructed with teachers and students as nodes. The edges of the dynamic interaction graph represent interaction relationships, and the weight of each edge is matched with the interaction intensity of the corresponding interaction relationship. The dynamic interaction graph is input into the spatiotemporal graph convolutional network, which extracts the spatiotemporal features of the interaction relationship to obtain the interaction feature vector of each node. We use a Long Short-Term Memory (LSTM) neural network to analyze the interaction feature vectors of each node and extract the interaction behavior features.

7. The public health teaching assessment and feedback system according to claim 6, characterized in that: The indicator parameter acquisition module determines indicator parameters based on teacher behavior characteristics, student behavior characteristics, and interaction behavior characteristics, including: Based on teacher behavior characteristics, student behavior characteristics, and interaction behavior characteristics, determine the distribution of student emotional state, classroom noise level, student attention span, student enthusiasm for answering questions, student participation behavior, teacher questioning frequency, student response rate, and duration of teacher-student interaction. The classroom atmosphere index is determined based on the distribution of students' emotional states, classroom noise levels, preset weights for emotional states, and preset weights for noise levels. The cumulative calculation method of time decay was used to analyze students' attention duration, students' enthusiasm for answering questions and students' participation behavior, and to determine the student participation index. The frequency of teacher questions, student response rate, and duration of teacher-student interaction were analyzed to determine indicators of teaching interaction effectiveness.

8. The public health teaching assessment and feedback system according to claim 1, characterized in that: The data acquisition module collects teaching outcome data and preprocesses it to obtain a teaching outcome dataset, including: Collect teaching outcome data including student attendance records, learning duration, frequency of classroom interaction, homework submission rate, homework accuracy rate, and test scores; The teaching outcome data is processed by denoising, missing value completion, outlier removal, and format quantization to obtain the teaching outcome dataset.

9. The public health teaching assessment and feedback system according to claim 8, characterized in that: The data analysis module performs in-depth analysis of the teaching outcome dataset to obtain teaching evaluation data, including: By integrating student attendance records, learning duration, and classroom interaction frequency, the single learning duration corresponding to each attendance and the classroom interaction frequency corresponding to each attendance are obtained. The single learning duration is corrected based on the classroom interaction frequency, and the average of the corrected single learning durations of all students is obtained to get the effective teaching duration. The homework submission rate, homework accuracy, and test scores are quantified to obtain homework completion rate and test accuracy. The homework completion rate and test accuracy are weighted to obtain knowledge mastery rate. The knowledge mastery rate of all students is estimated using a tracking model to obtain the teaching mastery rate.

10. The public health teaching assessment and feedback system according to claim 9, characterized in that: The second teaching quality assessment module evaluates teaching quality based on teaching assessment data, and obtains the second teaching quality assessment results, including: Compare the effective teaching time with the total pre-set course time to calculate the class completion rate index; Compare the teaching mastery with the pre-set knowledge mastery to calculate the knowledge attainment rate index; Based on the preset teaching objective weighting table, the class completion rate and knowledge attainment rate are weighted and integrated to calculate the comprehensive teaching effectiveness score. The comprehensive teaching effectiveness score is matched with the preset evaluation level range to obtain the second teaching quality assessment result.

Images