Self-adaptive teaching quality evaluation system based on machine learning

Abstract

The invention provides a self-adaptive teaching quality evaluation system based on machine learning. The system comprises an evaluation data collection unit, a classroom learning condition evaluation unit, a teacher teaching content evaluation unit and a teaching quality evaluation unit. The system objectively analyzes the teaching knowledge point coverage condition and the student input degree through the evaluation data collection unit, the classroom learning condition evaluation unit, the teacher teaching content evaluation unit and the teaching quality evaluation unit, comprehensively evaluates the classroom teaching quality in combination with a preset algorithm, improves the accuracy and comprehensiveness of the evaluation result, and improves the teaching quality. And powerful support is provided for teaching improvement.

Description

Technical Field

[0001] The present disclosure relates to the field of computer technology, and in particular to an adaptive teaching quality evaluation system based on machine learning. Background Art

[0002] At present, teaching quality assessment in related technologies mainly relies on data visualization statistical analysis technology, and specific methods include questionnaire surveys and student performance analysis.

[0003] However, questionnaires evaluate teaching quality by collecting feedback from students and teachers, which is greatly influenced by subjective factors and has low reference value. Although student grades can directly reflect students' learning effects, they often focus on the best and ignore the worst, making it difficult to fully reflect the teaching quality. Summary of the invention

[0004] The present invention provides an adaptive teaching quality assessment system based on machine learning.

[0005] According to a first aspect of the present disclosure, there is provided an adaptive teaching quality assessment system based on machine learning, the system comprising: an assessment data collection unit, a teacher teaching content assessment unit, a classroom learning situation assessment unit and a teaching quality assessment unit; the assessment data collection unit is used to collect teaching knowledge points contained in classroom teaching content and classroom behavior data during classroom teaching; the teacher teaching content assessment unit is used to evaluate the teaching situation of target important knowledge points in classroom teaching content based on the similarity between the teaching knowledge points and the target important knowledge point clusters, the target important knowledge point clusters are extracted from the textbook content and obtained after matching optimization; the classroom learning situation assessment unit is used to analyze the students' learning involvement based on the classroom behavior data to evaluate the students' performance in classroom teaching; the teaching quality assessment unit is used to evaluate the classroom teaching quality based on the teaching situation and performance according to a preset teaching quality assessment algorithm to obtain a classroom teaching quality assessment result.

[0006] In some embodiments, the teacher teaching content evaluation unit is used to: extract basic knowledge point clusters and important knowledge point clusters based on the textbook content, and optimize the target important knowledge point clusters based on the basic knowledge point clusters and the important knowledge point clusters; use a preset similarity evaluation algorithm, combined with the same-level matching rules, to evaluate the similarity of each level of knowledge points in the target important knowledge point cluster and the teaching knowledge point cluster of the teaching knowledge point in turn, and obtain knowledge point similarity results; and statistically analyze the teaching goal achievement degree evaluation indicators corresponding to the knowledge point similarity results.

[0007] In some embodiments, the important knowledge point clusters include a first important knowledge point cluster, a second important knowledge point cluster and a third important knowledge point cluster, and the teacher teaching content evaluation unit is used to: grade the textbook text knowledge points corresponding to the textbook content to obtain a basic knowledge point cluster; filter the first important knowledge point cluster based on the distribution of knowledge points in the basic knowledge point cluster; use a preset knowledge point screening model to filter qualified knowledge points from all knowledge points corresponding to the historical question bank to obtain a second important knowledge point cluster; use a hierarchical clustering algorithm and a correlation analysis algorithm to filter wrong question features under the target feature category from wrong question-related features, and use all knowledge points corresponding to the wrong question features as the third important knowledge point cluster; match and verify the first important knowledge point cluster based on the basic knowledge point cluster, the second important knowledge point cluster and the third important knowledge point cluster, and optimize the first important knowledge point cluster based on the matching verification result to obtain a target important knowledge point cluster.

[0008] In some embodiments, the teacher teaching content evaluation unit is used to: based on the same-level matching rule, use the second important knowledge point cluster and the third important knowledge point cluster to perform a first matching verification on all the first important knowledge points at each level in the first important knowledge point cluster in turn; if the first matching verification fails, then cross-matching verification is performed on each second important knowledge point in the second important knowledge point cluster and each third important knowledge point in the third important knowledge point cluster; if the cross-matching verification is successful, the cross-matching knowledge points that have successfully completed the cross-matching verification are determined, and based on the same-level matching rule, the cross-matching knowledge points are used to perform a second matching verification on all the basic knowledge points at each level in the basic knowledge point cluster in turn; if the second matching verification is successful, the matching knowledge points corresponding to the successful second matching verification are integrated into the first important knowledge point cluster, so that the first important knowledge point cluster is used as the target important knowledge point cluster.

[0009] In some embodiments, the teacher teaching content evaluation unit is used to: comprehensively obtain the matching verification result of each matching verification according to the first matching result obtained by the following formula 1 and the second matching result obtained by the following formula 2; Formula 1 in, is the first matching result between the knowledge point cluster G and the knowledge point cluster H, where the knowledge point cluster G and the knowledge point cluster H are any two knowledge point clusters that need to be matched and verified among the basic knowledge point cluster, the first important knowledge point cluster, the second important knowledge point cluster, and the third important knowledge point cluster. represents the number of knowledge points in the intersection of knowledge point cluster G and knowledge point cluster H, Represents the number of knowledge points in the union of knowledge point cluster G and knowledge point cluster H; Formula 2 in, is the second matching result between knowledge point cluster G and knowledge point cluster H, n is the nth knowledge point, N is the total number of knowledge points after knowledge point cluster G and knowledge point cluster H are aligned, is the nth knowledge point vector in the knowledge point cluster G, is the mean of the nth knowledge point vector after the knowledge point cluster G and the knowledge point cluster H are aligned, is the nth knowledge point vector in the knowledge point cluster H.

[0010] In some embodiments, classroom behavior data includes interactive behavior data and focused behavior data, and a classroom learning situation evaluation unit is used to: analyze students' learning engagement level under classroom teaching based on the interactive behavior data and / or focused behavior data corresponding to monitoring events triggered by classroom behavior operations under classroom teaching, combined with a preset behavior performance evaluation algorithm, to obtain students' classroom engagement level evaluation indicators, and monitoring events include interactive monitoring events and focused monitoring events.

[0011] In some embodiments, the evaluation data collection unit is used to: respond to interaction monitoring events triggered by classroom behavior operations under classroom teaching, collect teacher interaction audio and student interaction audio corresponding to the interaction monitoring events; identify teacher interaction topics corresponding to the teacher interaction audio and student interaction topics corresponding to the student interaction audio, and determine whether the teacher interaction topic matches the student interaction topic to obtain a matching result; based on the matching result, obtain interaction behavior data, the interaction behavior data including at least one of interaction duration, number of interaction failures, and number of interactions.

[0012] In some embodiments, the evaluation data collection unit is used to: respond to a concentration monitoring event triggered by a classroom behavior operation under classroom teaching, collect student classroom performance data under a classroom mode corresponding to the classroom teaching, the classroom mode is obtained by analyzing classroom instructions corresponding to the classroom behavior operation, the classroom mode includes a discussion mode and a concentration mode, and the student classroom performance data includes student audio and student facial features; based on the student audio and learning facial features, obtain concentration behavior data, the concentration behavior data includes at least one of the effective discussion duration, the invalid discussion duration, and the number of inattentive students.

[0013] In some embodiments, the system also includes: a student teaching experience evaluation unit, which is used to obtain the students' evaluation scores of the classroom teaching content contained in the classroom teaching content collected by the evaluation data collection unit; based on the number of scored students and each student's evaluation score of the classroom teaching content, combined with a preset satisfaction statistical algorithm, evaluate the students' satisfaction with the classroom teaching content, so as to evaluate the classroom teaching quality based on the teaching situation, satisfaction level and performance in accordance with the preset teaching quality evaluation algorithm, and obtain the classroom teaching quality evaluation results.

[0014] In some embodiments, according to a preset teaching quality evaluation algorithm, the teaching quality evaluation unit is used to: determine the classroom teaching quality evaluation result based on the teaching goal achievement evaluation index of the teaching situation and the student classroom engagement evaluation index of the student performance situation, combined with the preset teaching quality evaluation algorithm of the following formula 3; Formula 3 Among them, K is the result of classroom teaching quality evaluation, is the weighted coefficient of the mth evaluation index, is the mth evaluation indicator.

[0015] The present invention discloses an adaptive teaching quality evaluation system based on machine learning, comprising an evaluation data collection unit, a classroom learning situation evaluation unit, a teacher teaching content evaluation unit and a teaching quality evaluation unit; the evaluation data collection unit is used to collect teaching knowledge points contained in classroom teaching content and classroom behavior data in the classroom teaching process; the teacher teaching content evaluation unit is used to evaluate the teaching situation of target important knowledge points in classroom teaching content based on the similarity between the teaching knowledge points and the target important knowledge point clusters, and the target important knowledge point clusters are obtained after being extracted from the textbook content and matched and optimized; the classroom learning situation evaluation unit is used to analyze the learning input degree of students based on the classroom behavior data to evaluate the performance of students in classroom teaching; the teaching quality evaluation unit is used to evaluate the classroom teaching quality based on the teaching situation and performance according to the preset teaching quality evaluation algorithm, and obtain the classroom teaching quality evaluation result, which effectively solves the problems of inaccurate subjective evaluation and one-sided single performance evaluation in related technologies, and through objective analysis of the coverage of teaching knowledge points and the degree of student input, combined with the preset algorithm to comprehensively evaluate the classroom teaching quality, significantly improves the accuracy and comprehensiveness of the teaching evaluation results of classroom teaching, and provides strong support for teaching improvement.

[0016] It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present application, nor is it intended to limit the scope of the present application. Other features of the present application will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The accompanying drawings are used to better understand the present solution and do not constitute a limitation of the present disclosure. Figure 1 A schematic diagram of an adaptive teaching quality assessment system based on machine learning provided by an embodiment of the present disclosure; Figure 2 A flow chart of a teaching quality evaluation method applied to an adaptive teaching quality evaluation system provided by an embodiment of the present disclosure; Figure 3A flow chart of another teaching quality assessment method applied to an adaptive teaching quality assessment system provided by an embodiment of the present disclosure; Figure 4 A flow chart of another teaching quality assessment method applied to an adaptive teaching quality assessment system provided by an embodiment of the present disclosure; Figure 5 A schematic diagram of obtaining interactive behavior data provided by an embodiment of the present disclosure; Figure 6 A schematic diagram of obtaining focus behavior data provided by an embodiment of the present disclosure; Figure 7 A schematic diagram of a teaching scenario simulation module provided by an embodiment of the present disclosure; Figure 8 A structural schematic diagram of a teaching quality evaluation device applied to an adaptive teaching quality evaluation system provided by an embodiment of the present disclosure; Fig. 9 A schematic diagram of the hardware structure of an electronic device provided in an embodiment of the present disclosure. DETAILED DESCRIPTION

[0018] The following is a description of exemplary embodiments of the present disclosure in conjunction with the accompanying drawings, including various details of the embodiments of the present disclosure to facilitate understanding, which should be considered as merely exemplary. Therefore, it should be recognized by those of ordinary skill in the art that various changes and modifications may be made to the embodiments described herein without departing from the scope and spirit of the present disclosure. Similarly, for the sake of clarity and conciseness, descriptions of well-known functions and structures are omitted in the following description.

[0019] At present, teaching quality assessment in related technologies mainly relies on data visualization statistical analysis technology, and specific methods include questionnaire surveys and student performance analysis.

[0020] However, questionnaires evaluate teaching quality by collecting feedback from students and teachers, which is greatly influenced by subjective factors and has low reference value. Although student grades can directly reflect students' learning effects, they often focus on the best and ignore the worst, making it difficult to fully reflect the teaching quality.

[0021] It can be seen that the defects of instruction evaluation methods in related technologies are mainly reflected in two aspects. First, subjective evaluation methods such as questionnaires are easily affected by personal bias and emotions, resulting in inaccurate evaluation results. Second, student scores, as a single evaluation indicator, cannot fully reflect the quality of classroom teaching and ignore other important factors such as the teaching process and student involvement.

[0022] In order to solve the problem that it is difficult to comprehensively and objectively evaluate the teaching quality in related technologies, the adaptive teaching quality evaluation system based on machine learning proposed in the present invention can objectively evaluate the teaching situation of classroom teaching content on target important knowledge points by collecting teaching knowledge points and classroom behavior data in classroom teaching content, and at the same time analyze the students' learning input level to evaluate their performance in classroom teaching, and further comprehensively evaluate the teaching situation and students' performance according to the preset teaching quality evaluation algorithm, so as to accurately evaluate the quality of classroom teaching, thereby ensuring the accuracy and comprehensiveness of the quality evaluation and providing solid support for the continuous optimization of teaching quality.

[0023] The following describes, with reference to the accompanying drawings, an adaptive teaching quality assessment system based on machine learning, a teaching quality assessment method, an apparatus, an electronic device, a storage medium, and a computer program product applied to an adaptive teaching quality assessment system according to an embodiment of the present disclosure.

[0024] Figure 1 An adaptive teaching quality evaluation system based on machine learning is provided in the embodiments of the present disclosure, such as Figure 1 As shown, the system includes: Evaluation data collection unit 1, classroom learning assessment unit 2, teacher teaching content assessment unit 3 and teaching quality assessment unit 4; Evaluation data collection unit 1, used to collect teaching knowledge points contained in classroom teaching content and classroom behavior data during classroom teaching; Teacher teaching content evaluation unit 3, used to evaluate the teaching situation of target important knowledge points in classroom teaching content based on the similarity between teaching knowledge points and target important knowledge point clusters, where the target important knowledge point clusters are extracted from the textbook content and obtained after matching optimization; Classroom learning assessment unit 2, used to analyze students’ learning engagement based on classroom behavior data to evaluate students’ performance in classroom teaching; The teaching quality evaluation unit 4 is used to evaluate the classroom teaching quality based on the teaching situation and performance according to a preset teaching quality evaluation algorithm to obtain a classroom teaching quality evaluation result.

[0025] Among them, the system disclosed in the present invention also includes a student teaching experience evaluation unit, which is used to obtain the students' evaluation scores of the classroom teaching content contained in the classroom teaching content collected by the evaluation data collection unit; based on the number of scored students and each student's evaluation score of the classroom teaching content, combined with a preset satisfaction statistical algorithm, evaluate the students' satisfaction with the classroom teaching content, so as to evaluate the classroom teaching quality based on the teaching situation, satisfaction level and performance in accordance with the preset teaching quality evaluation algorithm, and obtain the classroom teaching quality evaluation results.

[0026] based on Figure 1 The embodiment shown, Figure 2 The following is a flow chart of a teaching quality evaluation method applied to an adaptive teaching quality evaluation system provided by an embodiment of the present disclosure. Figure 1 As shown, this method is applied to Figure 1 The adaptive teaching quality evaluation system in the invention comprises: Step 101, collecting teaching knowledge points contained in classroom teaching content and classroom behavior data during the classroom teaching process.

[0027] In some embodiments, the evaluation data collection unit of the present disclosure may specifically be teaching management. The evaluation data collection unit may collect and record specific teaching knowledge points covered in classroom teaching, which are usually directly related to the course content taught by the teacher. Secondly, the evaluation data collection unit may also capture and record various behavioral data of students during the course, which may include but are not limited to student participation (such as the number of times they raise their hands to speak, the activeness of group discussions), concentration (such as the frequency of eye contact with the teacher, the results of classroom quizzes), and classroom interaction performance (such as the number of questions asked, the accuracy of answering questions), etc.

[0028] In some embodiments, the assessment data collection unit may also integrate intelligent recognition technology (such as facial recognition, voice recognition, etc.) to collect students' behavioral data more accurately and without interference, ensuring the accuracy and objectivity of the data.

[0029] In an optional embodiment of the present disclosure, the present disclosure may obtain a cluster of teaching knowledge points from the course text of the course video corresponding to the classroom teaching by means of an evaluation data collection unit.

[0030] Specifically, the present disclosure can use IoT devices (such as high-definition cameras, smart blackboards, etc.) to record the entire classroom teaching process. These IoT devices can not only capture the real-time images of teachers teaching, but also record the teacher's explanation sound and the text content on the blackboard or screen. After the recording is completed, the video data will be safely and efficiently stored in the evaluation data collection unit for subsequent processing and analysis.

[0031] The evaluation data collection unit will use the pre-trained speech-to-text model to convert the audio content in the course video. Among them, the speech-to-text model in the present disclosure can use the Whisper model open sourced by OpenAI (or other similar models with superior performance) to accurately convert the speech content in the audio into text, providing a solid foundation for subsequent text analysis. Through the processing of the Whisper model (or similar models), the explanation content in the course video is efficiently converted into readable course text.

[0032] After obtaining the course text, the evaluation data collection unit can use text analysis technologies such as text mining, natural language, and semantic analysis to extract the teaching knowledge points in the course text, and grade the teaching knowledge points (for example, by chapter, difficulty, learning stage, etc.), encode the identification of each teaching knowledge point, and obtain the identification coding value corresponding to each teaching knowledge point, so as to construct a set of identification vectors corresponding to the teaching knowledge point identification based on the identification coding value, generate a teaching knowledge point network, and form all teaching knowledge point networks into the teaching knowledge point cluster.

[0033] In the present disclosure, the hierarchical identification can be divided into n levels. For example, when n=3, the teaching knowledge point network J=[Dn, Em, Fk], and the general expression is: J=(k, β), k is the knowledge point set, and β is the edge set.

[0034] When n=3, the teaching knowledge point network J in the present disclosure is a three-level classification system: the first-level classification Dn, the second-level classification Em and the third-level classification Fk. Each level of classification represents a level in the teaching knowledge point network, where the first-level classification is the main node, the second-level classification is the slave node of the first-level classification, and the third-level classification is the slave node of the second-level classification. That is, the first-level classification Dn is the main node, which can be classified by chapter, and Dn=[D1, D2...Dn] can be obtained; the second-level classification Em is the slave node of Dn, which can be classified by section, and Em=[D1e1, D1e2, D2e1..Dnem] can be obtained; the third-level classification Fk is the slave node of Em, which can be classified by unit results, and Fk=[D1e1f1, D1e2f2, D1e2f3..Dnemfk] can be obtained.

[0035] Because teaching knowledge points can appear in the form of a single unit or a combination of multiple units according to the degree of difficulty and learning depth, when constructing a teaching knowledge point network, it is determined that the first-level classification Dn exists in the form of a single-unit identification vector, and the second and third levels exist in the form of a combination of multiple units, with identification vectors superimposed in sequence.

[0036] The present invention forms a final teaching knowledge point cluster L=[Dn, Em, Fk] by combining multiple teaching knowledge point networks corresponding to the identification coding values ​​of the teaching knowledge points. The teaching knowledge point cluster of the present invention can cover all possible teaching knowledge point combinations, providing a rich material library for the push of questions.

[0037] In actual applications, the evaluation data collection unit can further intelligently select teaching knowledge points of different difficulty levels for push based on students' learning situations and needs. For example, for beginners or students who need to consolidate their basics, the evaluation data collection unit can push knowledge points of the first level classification; for students who want to learn in depth, the second or third level classification knowledge points can be pushed. This personalized push method not only improves learning efficiency, but also enhances students' interest and motivation in learning.

[0038] Step 102, based on the similarity between the teaching knowledge points and the target important knowledge point clusters, the teaching situation of the target important knowledge points in the classroom teaching content is evaluated. The target important knowledge point clusters are extracted from the teaching material content and obtained after matching optimization.

[0039] In some embodiments, the teacher teaching content evaluation unit of the present disclosure can first obtain the target important knowledge point cluster after matching optimization from the teaching material content, and the target important knowledge point cluster can represent the core learning objectives in the current teaching class and the key knowledge points that students are expected to master. The teacher teaching content evaluation unit of the present disclosure can quantitatively evaluate the coverage and explanation depth of the teaching content for the target knowledge points by calculating the similarity between the teaching knowledge points contained in the classroom teaching content and the target important knowledge point cluster, combined with a preset similarity evaluation algorithm, to obtain an evaluation index for the degree of achievement of the teaching objectives.

[0040] Among them, the teacher teaching content evaluation unit of the present disclosure needs to follow the same-level matching rule when determining the similarity between the teaching knowledge point cluster and the target important knowledge point cluster. That is, when comparing two knowledge points, it should be ensured that they are in the same knowledge level or classification, so as to avoid inappropriate comparison of knowledge points of different levels or types.

[0041] Step 103: Analyze students' learning engagement based on the classroom behavior data to evaluate students' performance in classroom teaching.

[0042] In some embodiments, the classroom learning situation evaluation unit of the present disclosure can obtain classroom behavior data collected in real time or regularly by the evaluation data collection unit. That is, the evaluation data collection unit can monitor and record various behaviors of students in the classroom, including interactive behaviors and focused behaviors. The collection of classroom behavior data is based on monitoring events triggered by classroom behavior operations. The classroom behavior data in the present disclosure may include interactive behavior data and focused behavior data.

[0043] Interactive behavior data mainly refers to the interactive behaviors between students and teachers or other students in class, such as asking questions, answering, discussing, group cooperation, etc., which can reflect the students' participation and enthusiasm.

[0044] Focus behavior data mainly focuses on students' concentration in class, such as whether they listen carefully, whether they take notes, whether they are often distracted, etc., which can reflect students' learning attitude and level of involvement.

[0045] The classroom learning situation evaluation unit disclosed in the present invention can use a preset behavior performance evaluation algorithm to comprehensively consider the interactive behavior data and the focused behavior data (or one of them), as well as their weights, to calculate a comprehensive learning engagement score, that is, a student classroom engagement evaluation index, to evaluate students' performance in classroom teaching.

[0046] Step 104, according to a preset teaching quality evaluation algorithm, the classroom teaching quality is evaluated based on the teaching situation and performance, and a classroom teaching quality evaluation result is obtained.

[0047] In some embodiments, the teaching quality evaluation unit of the present disclosure may determine the classroom teaching quality evaluation result based on the teaching goal achievement degree evaluation index of the teaching situation and the student classroom engagement degree evaluation index of the student performance situation, combined with the preset teaching quality evaluation algorithm of the following formula 3; Formula 3 Among them, K is the result of classroom teaching quality evaluation, is the weighted coefficient of the mth evaluation index, is the mth evaluation indicator.

[0048] Among them, in order to further improve the accuracy of the classroom teaching quality evaluation results, the present disclosure can also collect the students' evaluation scores of the classroom teaching content contained in the classroom teaching content through the evaluation data collection unit. The evaluation data collection unit obtains the students' evaluation scores of the classroom teaching content, and based on the number of scored students and the evaluation scores of each student on the classroom teaching content, combined with a preset satisfaction statistical algorithm, evaluates the students' satisfaction with the classroom teaching content, so as to evaluate the classroom teaching quality based on the teaching situation, satisfaction and performance according to the preset teaching quality evaluation algorithm, and obtain the classroom teaching quality evaluation results.

[0049] Specifically, the present disclosure may collect the evaluation scores of students on classroom teaching content contained in the teaching content fed back by students through an evaluation data collection unit, wherein the evaluation subject may be a student, or a teacher or other relevant personnel, which is not limited in the present disclosure embodiment.

[0050] After obtaining the evaluation scores, the student teaching experience evaluation unit of the present invention can evaluate students' satisfaction with the classroom teaching content based on the preset satisfaction statistical algorithm of the following formula, based on the number of scored students and each student's evaluation score of the classroom teaching content, to obtain a classroom satisfaction evaluation index.

[0051]

[0052] in, is the classroom satisfaction evaluation index, q is the number of scoring students (specifically, it can be the total number of satisfaction questionnaires collected), Hi is the evaluation score of the i-th scoring student on the classroom teaching content (the evaluation score can be specifically the satisfaction score), It is the basic score for the third teaching situation evaluation indicator.

[0053] The teaching quality evaluation unit disclosed in the present invention can determine the classroom teaching quality evaluation result based on the preset teaching quality evaluation algorithm of the above formula 3 according to the obtained classroom satisfaction evaluation index, teaching goal achievement evaluation index and student classroom involvement evaluation index.

[0054] In summary, the technical solution provided by the present invention collects teaching knowledge points contained in classroom teaching content and classroom behavior data in the classroom teaching process; evaluates the teaching situation of target important knowledge points in classroom teaching content based on the similarity between teaching knowledge points and target important knowledge point clusters, and the target important knowledge point clusters are extracted from the textbook content and obtained after matching and optimization; analyzes students' learning engagement based on classroom behavior data to evaluate students' performance in classroom teaching; evaluates classroom teaching quality based on teaching situation and performance according to a preset teaching quality evaluation algorithm, and obtains classroom teaching quality evaluation results, which effectively solves the problems of inaccurate subjective evaluation and one-sided single score evaluation in related technologies, and objectively analyzes the coverage of teaching knowledge points and students' engagement, and comprehensively evaluates classroom teaching quality in combination with a preset algorithm, which significantly improves the accuracy and comprehensiveness of the evaluation results, and provides strong support for teaching improvement.

[0055] As a possible implementation, Figure 3 The flowchart of another adaptive teaching quality evaluation system based on machine learning is shown. On the basis of the above embodiment, the specific process of evaluating the teaching situation of the target important knowledge points in the classroom teaching content based on the similarity between the teaching knowledge points and the target important knowledge point clusters includes the following steps, which are implemented by the teacher teaching content evaluation unit: Step 201 , extracting basic knowledge point clusters and important knowledge point clusters based on the textbook content, and optimizing the basic knowledge point clusters and important knowledge point clusters to obtain target important knowledge point clusters.

[0056] In some embodiments, the important knowledge point clusters include a first important knowledge point cluster, a second important knowledge point cluster and a third important knowledge point cluster. The teacher teaching content evaluation unit can extract the basic knowledge point cluster and the important knowledge point cluster based on the textbook content, and optimize the target important knowledge point cluster based on the basic knowledge point and the important knowledge point cluster, including: grading the textbook text knowledge points corresponding to the textbook content to obtain the basic knowledge point cluster; screening the first important knowledge point cluster based on the distribution of knowledge points in the basic knowledge point cluster; using a preset knowledge point screening model to screen qualified knowledge points from all knowledge points corresponding to the historical question bank to obtain the second important knowledge point cluster; using a hierarchical clustering algorithm and a correlation analysis algorithm to screen the wrong question features under the target feature category from the wrong question related features, and use all knowledge points corresponding to the wrong question features as the third important knowledge point cluster; matching and verifying the first important knowledge point cluster based on the basic knowledge point cluster, the second important knowledge point cluster and the third important knowledge point cluster, and optimizing the first important knowledge point cluster based on the matching verification result to obtain the target important knowledge point cluster.

[0057] In an optional embodiment of the present disclosure, for the acquisition of basic knowledge point clusters, the teacher teaching content evaluation unit of the present disclosure can use text analysis technologies such as text mining, natural language, and semantic analysis to extract textbook knowledge points in the textbook text, and perform classification processing on the textbook text knowledge points according to preset classification rules (for example, according to chapters) to obtain basic knowledge point clusters. Among them, the classification step can specifically adopt the classification processing method for the teaching knowledge point cluster in the above step 101, which will not be repeated in the embodiment of the present disclosure. The present disclosure can use the above classification processing method to obtain the basic knowledge point cluster Uori=[An, Bm, Ck].

[0058] In an optional embodiment of the present disclosure, in order to obtain the first important knowledge point cluster, the teacher teaching content evaluation unit of the present disclosure can count the number of basic knowledge points at each level in the basic knowledge cluster, and determine the basic knowledge cluster score based on the number of basic knowledge points at each level according to the following formula, so as to screen and obtain the first important knowledge point cluster according to the basic knowledge point cluster scoring results.

[0059]

[0060] in, is the scoring result of the basic knowledge point cluster; Z is the number of basic knowledge points at the first level in the basic knowledge point cluster; X is the number of basic knowledge points at the second level in the basic knowledge point cluster; V is the number of basic knowledge points at the third level in the basic knowledge point cluster; M is the number of basic knowledge points at the nth level in the basic knowledge point cluster; is the weight of the number of basic knowledge points at the first level, is the weight of the number of basic knowledge points at the second level, is the weight of the number of basic knowledge points at the third level; is the weight of the number of basic knowledge points at the nth level.

[0061] in, < < , because as the knowledge points become more refined, their importance accounts for a larger proportion.

[0062] After obtaining the basic knowledge point cluster scoring result, the present disclosure can compare the basic knowledge point cluster scoring result Q with the preset screening threshold T1. When Q is greater than or equal to T1, the knowledge point identification vector in the screened basic knowledge point cluster is used as the first important knowledge point cluster Unew. Among them, the preset screening threshold can be set by custom, average method, median, etc., which is not limited in the embodiment of the present disclosure.

[0063] In an optional embodiment of the present disclosure, in order to obtain the second important knowledge point cluster, the teacher teaching content evaluation unit of the present disclosure can perform graded processing on all knowledge points extracted from the history question bank text to obtain the history question bank knowledge point cluster (wherein, the grading step can specifically adopt the grading processing method for the teaching knowledge point cluster in the above step 101, which will not be repeated in the embodiment of the present disclosure. The present disclosure can use the above-mentioned grading processing method to obtain the history question bank knowledge point cluster Zori=[Wn, Rm, Hk]).

[0064] The teacher teaching content evaluation unit disclosed in the present invention can determine the knowledge point score of each knowledge point in the history question bank knowledge point cluster according to the preset knowledge point screening model of the following formula, so as to screen the qualified knowledge points from the knowledge point score of each knowledge point to obtain the second important knowledge point cluster.

[0065]

[0066] in, Score the knowledge point P in the knowledge point cluster of the history question bank, is the number of occurrences of knowledge point P, is the number of errors in knowledge point P, is the number of knowledge points at the level corresponding to knowledge point P, is the weight corresponding to the number of occurrences, The weight corresponding to the number of errors, is the weight corresponding to the number of knowledge points.

[0067] After obtaining the knowledge point score of each knowledge point, the teacher teaching content evaluation unit of the present disclosure can compare the knowledge point score of each knowledge point. and the preset scoring threshold T2, when When it is greater than or equal to T2, the important knowledge point combinations are screened out to form the second important knowledge point cluster Znew.

[0068] In an optional embodiment of the present disclosure, in order to obtain the third important knowledge point cluster, the teacher teaching content evaluation unit of the present disclosure can obtain the wrong question correlation characteristics of the subjective and objective factors that affect students' wrong answers collected by the evaluation data collection unit, and use cluster analysis and correlation evaluation models to screen out the wrong question characteristics caused by objective factors, and then determine the feature group or features with the greatest impact, and accordingly construct the third important knowledge point cluster Onew that focuses on students' mastery of knowledge points, so as to accurately improve students' learning outcomes.

[0069] Specifically, the present disclosure can collect factors that affect wrong answers through the evaluation data collection unit, including emotions, concentration, answering style, knowledge point mastery, answering efficiency, question difficulty level, etc., and simulate the test scene through the evaluation data collection unit to collect students' answering video screens, and obtain subjective factor feature data sets through Single Shot Multibox Detector (SSD) facial expression recognition technology and eye tracking technology; collect students' answering situations through the evaluation data collection unit to obtain objective factor feature data sets. After obtaining the above-mentioned wrong question related features, the evaluation data collection unit can feed back the wrong question related features to the teacher teaching content evaluation unit.

[0070] Among them, subjective factors include: emotional characteristics (such as positive, indifferent, negative, etc.), concentration characteristics (such as concentration, distraction, susceptibility to external factors, etc.), test-taking style characteristics (such as carefulness, carelessness, lack of patience, etc.). Objective factors include: knowledge point mastery characteristics (such as proficiency, proficiency, unproficiency, etc.), test-taking efficiency characteristics (such as fast, medium, slow, etc.), and question difficulty level characteristics (such as simple, ordinary, difficult, etc.).

[0071] The teacher teaching content evaluation unit of the present disclosure can use clustered wrong question related feature groups to analyze the influence of different groups of wrong question related features on wrong questions. That is, the present disclosure can use random combination of all wrong question related features to find out the objective factors that really affect wrong questions. This is because there is data interference from subjective factors, and it is impossible to accurately judge whether students have mastered the corresponding knowledge points. Therefore, the present disclosure needs to screen out the wrong question feature data set caused by objective factors from the wrong question related feature data set, thereby improving the accuracy of the third important knowledge point cluster.

[0072] Among them, when performing cluster analysis, the present disclosure needs to first perform data preprocessing on the acquired wrong question related features, such as processing missing values ​​and outliers for the wrong question related features to ensure data quality, and perform Z-score standardization on the wrong question related features.

[0073] Afterwards, because of the difficulty of the questions, the efficiency of answering questions is slow, which is a normal phenomenon. Therefore, the teacher teaching content evaluation unit disclosed in the present invention needs to first use a hierarchical clustering algorithm to group the pre-processed wrong question related features, and use the Pearson correlation coefficient distance algorithm of the following formula to calculate the distance matrix between each wrong question related feature to obtain the first correlation analysis result.

[0074]

[0075] in, is the result of the first correlation analysis, , is the average value of wrong question related features a and wrong question related features b, where a and b are wrong question related features. The value range is [0, 1]. The closer the value is to 1, the looser the linear correlation between the wrong question association feature and the wrong question feature.

[0076] Based on the above first correlation analysis results, hierarchical clustering is selected to gradually merge or split clusters to generate a dendrogram to obtain the correlation distribution between the associated features of wrong questions.

[0077] For example, the present disclosure may select the shortest distance linking method: d(Ccluster,Dcluster)=min{d(c,d):c∈C,d∈D}, where d(Ccluster,Dcluster) represents the distance between cluster C and cluster D, and d(Ccluster,Dcluster) represents the distance between a point c in cluster C and a point d in cluster D. Determine the number of groups k of wrong question-related features, and assign the wrong question-related features to the corresponding groups to obtain the correlation distribution between the wrong question-related features.

[0078] The teacher teaching content evaluation unit of the present disclosure analyzes each set of wrong question related features, evaluates the relationship between them and the wrong question features again, and obtains the second correlation analysis result (i.e., the correlation distribution between the wrong question related feature sets is generated based on the second correlation analysis result, and whether the wrong question related feature sets have correlation) through the pre-trained correlation evaluation model: The relevance evaluation model is as follows:

[0079] Among them, y is the result of the second correlation analysis, is the intercept, is the partial regression coefficient, and x1, x2...xn are the associated features of wrong questions.

[0080] The present disclosure selects the wrong question-associated feature group or feature that has the greatest impact on wrong questions through the first correlation analysis result and the second correlation analysis result (i.e., selects the wrong question-associated feature with the maximum mean value of the first correlation analysis result and the second correlation analysis result of each wrong question-associated feature, or selects the wrong question-associated feature with the maximum value of the weighted fusion of the first correlation analysis and the second correlation analysis result of each wrong question-associated feature), and determines the feature category of each screened wrong question-associated feature (feature categories include subjective categories and objective categories). The present disclosure uses the wrong question-associated feature whose feature category is the target feature category (i.e., objective category) as the wrong question feature, thereby constructing the third important knowledge point cluster Onew.

[0081] Among them, the subjective category reflects the students' own psychology, not the reason for making mistakes due to unfamiliarity with knowledge points. What needs to be improved is their own psychological quality; the objective category reflects that students make mistakes due to insufficient knowledge points. Because the subjective category requires students to adjust themselves or a third party to assist in guidance, and only the objective category is the essential reason for making mistakes due to insufficient knowledge points, so the objective category is used as the target feature category.

[0082] In some embodiments, after obtaining the basic knowledge point cluster, the first important knowledge point cluster, the second important knowledge point cluster and the third important knowledge point cluster, the teacher teaching content evaluation unit of the present disclosure can also optimize the first important knowledge point cluster. Specifically, it includes: based on the same-level matching rule, using the second important knowledge point cluster and the third important knowledge point cluster, first matching verification is performed on all the first important knowledge points at each level in the first important knowledge point cluster in turn; if the first matching verification fails, cross-matching verification is performed on each second important knowledge point in the second important knowledge point cluster and each third important knowledge point in the third important knowledge point cluster; if the cross-matching verification is successful, the cross-matching knowledge points that have successfully been cross-matched are determined, and based on the same-level matching rule, the cross-matching knowledge points are used to perform a second matching verification on all the basic knowledge points at each level in the basic knowledge point cluster in turn; if the second matching verification is successful, the matching knowledge points corresponding to the second matching verification success are integrated into the first important knowledge point cluster, so as to use the first important knowledge point cluster as the target important knowledge point cluster.

[0083] Among them, in the matching verification disclosed in the present invention, if the matching verification of the current level fails, the matching verification of the next level will be automatically performed, until all current levels have been matched and verified, and different operations will be performed according to the matching verification results. The cross-matching in the present invention refers to the matching verification of the same level matching between the second knowledge point cluster and the third knowledge point cluster after the first matching verification of the first important knowledge point cluster fails, to determine whether there are matching knowledge points in the second knowledge point cluster and the third knowledge point cluster, so as to avoid the situation of knowledge point recognition errors caused by system errors.

[0084] If the first matching verification is successful, the successfully matched knowledge point is bound to the question bank and imported into the first key knowledge point cluster. If the cross matching verification fails or the second matching verification fails, the knowledge point cluster currently undergoing matching verification is stored in the candidate library.

[0085] Specifically, the present disclosure may obtain the matching verification result of each matching verification comprehensively according to the first matching result obtained by the following formula 1 and the second matching result obtained by the following formula 2; Formula 1 in, is the first matching result between the knowledge point cluster G and the knowledge point cluster H, where the knowledge point cluster G and the knowledge point cluster H are any two knowledge point clusters that need to be matched and verified among the basic knowledge point cluster, the first important knowledge point cluster, the second important knowledge point cluster, and the third important knowledge point cluster. represents the number of knowledge points in the intersection of knowledge point cluster G and knowledge point cluster H, Represents the number of knowledge points in the union of knowledge point cluster G and knowledge point cluster H; Formula 2 in, is the second matching result between knowledge point cluster G and knowledge point cluster H, n is the nth knowledge point, N is the total number of knowledge points after knowledge point cluster G and knowledge point cluster H are aligned, is the nth knowledge point vector in the knowledge point cluster G, is the mean of the nth knowledge point vector after the knowledge point cluster G and the knowledge point cluster H are aligned, is the nth knowledge point vector in the knowledge point cluster H.

[0086] In the present disclosure, when the first match verification result is greater than or equal to the first match verification threshold, and the second match verification result is greater than the second match verification threshold, the current match verification result is determined to be a successful match.

[0087] It is understandable that since the verification matching in the present disclosure adopts the principle of same-level matching, if only relying on the matching verification algorithm of formula 1 (for example, the Jaccard similarity algorithm), you may encounter a situation: even if the matching degree of two objects seems similar overall (such as the output result of q (1, 0, 1)), there is a mismatch in the more detailed secondary dimension (such as the secondary dimension value is 0), which does not meet the consistency standard. Therefore, in order to make up for this deficiency, the present disclosure further combines the matching verification algorithm of formula 2 (for example, the improved cosine similarity algorithm) for verification. The improved cosine similarity algorithm is better than other algorithms in measuring the difference in the values ​​of each dimension, but its accuracy still has certain limitations. Therefore, by combining these two algorithms, the present disclosure can achieve complementary advantages, thereby improving the accuracy and reliability of matching.

[0088] Step 202, using a preset similarity evaluation algorithm and combining the same-level matching rules, the similarity evaluation is performed on the knowledge points of each level in the target important knowledge point cluster and the teaching knowledge point cluster of the teaching knowledge point in turn to obtain the knowledge point similarity result.

[0089] In some embodiments, the teacher teaching content evaluation unit of the present invention uses the same-level matching rules to evaluate the similarity of the knowledge points of each level in the target important knowledge point cluster and the teaching knowledge point cluster of the teaching knowledge point in turn. Specifically, it can be achieved through the content of the same-level matching in the above step 201, wherein the similarity evaluation between the knowledge point clusters is specifically the matching verification between every two knowledge point clusters in the above-mentioned same-level matching process. The preset similarity evaluation algorithm in the present invention is a combination of the above-mentioned formula 1 and formula 2 (that is, the first similarity and the second similarity between every two knowledge point clusters are determined respectively by the above-mentioned formula 1 and formula 2. If the first similarity is greater than or equal to the first similarity threshold, and the second similarity is greater than or equal to the second similarity threshold, it is determined that the knowledge point similarity result between the current target important knowledge point cluster and the teaching knowledge point cluster is similar), and the specific process will not be repeated here.

[0090] Step 203, counting the teaching goal achievement degree evaluation indicators corresponding to the knowledge point similarity results.

[0091] In some embodiments, after obtaining the knowledge point similarity results between the target important knowledge point cluster and the teaching knowledge point cluster, the teacher teaching content evaluation unit of the present invention can calculate the teaching objective achievement degree evaluation index corresponding to the knowledge point similarity results based on the preset similarity statistical algorithm of the following formula. The teaching objective achievement degree evaluation index may refer to the hit score of the teaching content for the target important knowledge points.

[0092]

[0093] Among them, D1 is the evaluation index of the degree of achievement of the teaching goal, m is the total number of similarity evaluations that have been conducted, p is the number of evaluations with similarity, Pn is the similarity value between the nth knowledge points (specifically, it can be the target similarity value after weighted fusion of the first similarity and the second similarity, n is the knowledge point with similarity), i is the total number of similarity evaluations required; Pm is the mth similarity evaluation value (specifically, it can be the target similarity value after weighted fusion of the first similarity and the second similarity); It is the basic score for evaluating the degree of achievement of teaching objectives.

[0094] In summary, the present disclosure accurately extracts the basic knowledge point clusters and important knowledge point clusters in the teaching material content through the teacher teaching content evaluation unit, and optimizes the target important knowledge point clusters, combines the preset similarity evaluation algorithm and the same-level matching rules, and conducts a comprehensive and objective similarity evaluation of the knowledge points at each level, thereby statistically calculating the teaching goal achievement degree evaluation index. This not only improves the accuracy and comprehensiveness of the evaluation, but also provides strong data support for the improvement of teaching quality and the optimization of student learning effects.

[0095] As a possible implementation, Figure 4 The flowchart of another adaptive teaching quality evaluation system based on machine learning is shown. Based on the above embodiment, the classroom behavior data includes interactive behavior data and focused behavior data. The specific process of analyzing the students' learning engagement based on the classroom behavior data to evaluate the students' performance in classroom teaching includes the following steps, which are implemented by the classroom learning situation evaluation unit: Step 301, based on the interactive behavior data and / or focused behavior data corresponding to the monitoring events triggered by classroom behavior operations under classroom teaching, combined with a preset behavior performance evaluation algorithm, analyze the students' learning engagement level under classroom teaching to obtain the students' classroom engagement level evaluation index. The monitoring events include interactive monitoring events and focused monitoring events.

[0096] In some embodiments, the classroom learning situation evaluation unit of the present disclosure can be based on the interactive behavior data and / or focused behavior data in the teaching classroom obtained by the evaluation data collection unit; based on the interactive behavior data and / or focused behavior data, combined with a preset behavior performance evaluation algorithm, evaluate the student learning situation in the teaching classroom, and obtain the student classroom engagement evaluation index, that is, the learning behavior performance evaluation result. Among them, the evaluation data collection unit obtains the interactive behavior data and / or focused behavior data in the teaching classroom based on the monitoring events triggered by the teacher's behavior operations and student's behavior operations in the teaching classroom, and the monitoring events include interactive monitoring events triggered by the teacher's dialogue operation and the student's dialogue operation at the same time and the concentration monitoring event triggered by the teacher's instruction issuance operation.

[0097] In some embodiments, the evaluation data collection unit of the present disclosure can respond to interaction monitoring events triggered by classroom behavior operations under classroom teaching, collect teacher interaction audio and student interaction audio corresponding to the interaction monitoring events; identify the teacher interaction topic corresponding to the teacher interaction audio and the student interaction topic corresponding to the student interaction audio, and determine whether the teacher interaction topic matches the student interaction topic to obtain a matching result; based on the matching result, obtain interaction behavior data, the interaction behavior data including at least one of the interaction duration, the number of interaction failures, and the number of interactions.

[0098] Specifically, Figure 5 A schematic diagram of obtaining interactive behavior data is shown. The evaluation data collection unit of the present disclosure can respond to interactive monitoring events triggered simultaneously by the teacher's dialogue operation on the podium central console and the student's dialogue operation on the student-side interactive device in the teaching classroom, and obtain the teacher interaction audio of the teacher in the teaching classroom and the student interaction audio of the first interactive student, where the first interactive student is determined according to the trigger time of each interactive student corresponding to the student dialogue operation; based on a preset topic recognition model (i.e., the LDA topic model), the teacher interaction topic corresponding to the teacher interaction audio and the student interaction topic corresponding to the student interaction audio are respectively identified; if the teacher interaction topic is the same as the student interaction topic, the interaction time corresponding to the student interaction audio is counted; if the teacher interaction topic is different from the student interaction topic, it is determined that the interaction of the first interactive student has failed, and based on the interaction failure result, the number of interaction failures in the teaching classroom is counted.

[0099] Among them, for the number of interactions, the evaluation data collection unit of the present invention can sort all interactive students according to each trigger time of each interactive student corresponding to the student dialogue operation to obtain the sorting result; determine the first interactive student who ranks first in the sorting result; count the total number of student dialogue operations among all interactive students except the first interactive student, and use the total number of student dialogue operations as the number of interactions.

[0100] The evaluation data collection unit disclosed in the present invention can also respond to a concentration monitoring event triggered by a classroom behavior operation under classroom teaching, and collect student classroom performance data under a classroom mode corresponding to the classroom teaching. The classroom mode is obtained by analyzing classroom instructions corresponding to the classroom behavior operation. The classroom mode includes a discussion mode and a concentration mode. The student classroom performance data includes student audio and student facial features. Based on the student audio and learning facial features, the concentration behavior data is obtained, and the concentration behavior data includes at least one of the effective discussion duration, the invalid discussion duration, and the number of inattentive students.

[0101] Specifically, Figure 6A schematic diagram of obtaining concentration behavior data is shown. The evaluation data collection unit of the present disclosure can respond to a concentration monitoring event triggered by a teacher issuing a teacher instruction through a teacher's instruction issuing operation on a podium console in a teaching classroom, and determine the course mode corresponding to the teacher instruction of the teacher instruction issuing operation, and the classroom mode includes a discussion mode and a concentration mode; when the classroom mode is the discussion mode, the discussion audio of all students in the teaching classroom is obtained, and based on the audio decibel of the discussion audio and a preset decibel threshold, the student discussion situation in the teaching classroom is analyzed to determine the effective discussion time and the invalid discussion time corresponding to the discussion audio; when the classroom mode is the concentration mode, the conversation audio and facial expression features of all students in the teaching classroom are obtained, and based on the conversation audio and facial expression features, the abnormal behavior of students in the teaching classroom is analyzed to determine the number of inattentive students who have abnormal learning behavior.

[0102] Among them, based on the conversation audio and facial expression features, the abnormal behaviors of students in the teaching classroom are analyzed, and the number of inattentive students with abnormal learning behaviors is determined, including: determining the conversation duration corresponding to the conversation audio of each student in the teaching classroom, and based on the conversation duration, counting the number of first students with abnormal conversation behaviors in the teaching classroom, where the abnormal conversation behavior includes a conversation duration greater than a preset duration threshold, a conversation duration less than or equal to the preset duration threshold, and the number of conversations greater than a preset conversation number threshold; analyzing the facial expression changes corresponding to the facial expression features of each student in the teaching classroom, and based on the facial expression changes, counting the number of second students with inattentive facial expressions; merging the number of first students and the number of second students to obtain the number of inattentive students.

[0103] After obtaining the first number of students and the second number of students, the present invention can also feed back the obtained first number of students and the second number of students to the podium console, so that the teacher can view the current students' classroom performance in real time.

[0104] Further, after obtaining the interactive behavior data and / or focused behavior data fed back by the evaluation data collection unit, the classroom learning situation evaluation unit of the present disclosure can analyze the learning engagement of students in classroom teaching based on the interactive behavior data and focused behavior data according to the preset behavior performance evaluation algorithm of the following formula 3, and obtain the evaluation index of the students' classroom engagement;

[0105] in, is the evaluation index of students’ classroom involvement, S is the number of interactions in the interactive behavior data, is the preset interaction threshold. is the interaction duration in the interactive behavior data, is the total number of hours of teaching class. is the effective discussion duration in the focus behavior data, C is the number of interaction failures in the interaction behavior data, is the preset interaction failure threshold, A is the number of inattentive students in the attention behavior data, is the total number of students in the teaching class, To focus on the ineffective discussion time in behavioral data, , , , , as well as They are the weight values ​​between the ratios of each data in the interactive behavior data and the focused behavior data.

[0106] In addition, in the present disclosure, obtaining interactive behavior data and focused behavior data and obtaining student classroom engagement evaluation indicators based on the interactive behavior data and focused behavior data can be achieved through an evaluation data collection unit, that is, through a teaching scenario simulation module in the evaluation data collection unit.

[0107] Specific as Figure 7 The schematic diagram of a teaching scene simulation module shown in the figure can include a podium central console, a smart blackboard, a sound collection device, and a student-side interactive device. The smart blackboard is used to synchronize course content and is interconnected with the podium central console; the podium central console is used for course presentations, interaction with student-side interactive devices, data processing, and other operations; the sound collection device is used to collect sounds in the classroom; the student-side interactive device: used to monitor student status, participate in classroom interaction, data collection, and other operations.

[0108] The present invention can receive classroom instructions issued by the podium console through a sound collection device, and collect classroom discussion audio according to the classroom instructions, analyze the effective discussion time and the invalid discussion time, and feed back the effective discussion time and the invalid discussion time to the podium console.

[0109] The student-side interactive device in the present disclosure can receive classroom instructions issued by the podium console, obtain students' classroom conversation audio and classroom learning expression features, and analyze the classroom conversation audio and classroom learning expression features to obtain the number of inattentive students, so as to feed back the number of inattentive students to the podium console; The podium console disclosed in the present invention obtains the interaction duration, the number of interaction failures and the number of interactions based on the student interaction audio of the student interaction events fed back by the student-side interactive device, so as to integrate and analyze all the classroom concentration data and classroom interaction data to obtain the student classroom engagement level evaluation index according to the above formula.

[0110] In summary, the present invention can accurately capture students' interactive behavior data and focused behavior data by real-time monitoring of various monitoring events triggered by teachers' and students' behavioral operations in the teaching classroom, including interactive monitoring events (such as teacher-student dialogues) and concentration monitoring events (such as students' reactions after the teacher's instructions are issued), and combine them with a preset behavioral performance evaluation algorithm to comprehensively evaluate students' learning engagement and learning situation, and then derive students' classroom engagement evaluation indicators, which not only improves the accuracy and objectivity of teaching evaluation, but also helps teachers to adjust teaching strategies in a timely manner and optimize classroom interaction, thereby effectively improving students' learning outcomes.

[0111] Corresponding to the teaching quality assessment method of the above-mentioned adaptive teaching quality assessment system based on machine learning, the present invention also proposes a teaching quality assessment device applied to the adaptive teaching quality assessment system. Since the device embodiment of the present invention corresponds to the above-mentioned method embodiment, the details not disclosed in the device embodiment can be referred to the above-mentioned method embodiment, and will not be repeated in the present invention.

[0112] Figure 8 A structural diagram of a teaching quality evaluation device applied to an adaptive teaching quality evaluation system provided by an embodiment of the present disclosure is shown in FIG. Figure 8 As shown, the device comprises: The data collection unit 610 is used to collect the teaching knowledge points contained in the classroom teaching content and the classroom behavior data in the classroom teaching process; A teaching content evaluation unit 620 is used to evaluate the teaching of target important knowledge points in the classroom teaching content based on the similarity between the teaching knowledge points and the target important knowledge point clusters, where the target important knowledge point clusters are extracted from the teaching material content and then matched and optimized; A student performance evaluation unit 630 is used to analyze the student's learning engagement based on the classroom behavior data to evaluate the student's performance in classroom teaching; The teaching quality evaluation unit 640 is used to evaluate the classroom teaching quality based on the teaching situation and performance according to a preset teaching quality evaluation algorithm to obtain a classroom teaching quality evaluation result.

[0113] In some embodiments of the present disclosure, the teaching content evaluation unit 620 is used to: extract basic knowledge point clusters and important knowledge point clusters based on the textbook content, and optimize the target important knowledge point clusters based on the basic knowledge point clusters and the important knowledge point clusters; use a preset similarity evaluation algorithm, combined with the same-level matching rules, to sequentially evaluate the similarity of the target important knowledge point clusters and the teaching knowledge point clusters of the teaching knowledge points at each level to obtain knowledge point similarity results; and statistically calculate the teaching objective achievement degree evaluation indicators corresponding to the knowledge point similarity results.

[0114] In some embodiments of the present disclosure, the important knowledge point clusters include a first important knowledge point cluster, a second important knowledge point cluster and a third important knowledge point cluster, and the teaching content evaluation unit 620 is used to: perform hierarchical processing on the textbook text knowledge points corresponding to the textbook content to obtain a basic knowledge point cluster; filter out the first important knowledge point cluster based on the distribution of knowledge points in the basic knowledge point cluster; use a preset knowledge point screening model to filter qualified knowledge points from all knowledge points corresponding to the historical question bank to obtain a second important knowledge point cluster; use a hierarchical clustering algorithm and a correlation analysis algorithm to filter out wrong question features under the target feature category from wrong question-related features, and use all knowledge points corresponding to the wrong question features as the third important knowledge point cluster; match and verify the first important knowledge point cluster based on the basic knowledge point cluster, the second important knowledge point cluster and the third important knowledge point cluster, and optimize the first important knowledge point cluster based on the matching verification result to obtain a target important knowledge point cluster.

[0115] In some embodiments of the present disclosure, the teaching content evaluation unit 620 is used to: based on the same-level matching rule, use the second important knowledge point cluster and the third important knowledge point cluster to perform a first matching verification on all the first important knowledge points at each level in the first important knowledge point cluster in turn; if the first matching verification fails, then cross-matching verification is performed on each second important knowledge point in the second important knowledge point cluster and each third important knowledge point in the third important knowledge point cluster; if the cross-matching verification is successful, the cross-matching knowledge points that have successfully completed the cross-matching verification are determined, and based on the same-level matching rule, the cross-matching knowledge points are used to perform a second matching verification on all the basic knowledge points at each level in the basic knowledge point cluster in turn; if the second matching verification is successful, the matching knowledge points corresponding to the successful second matching verification are integrated into the first important knowledge point cluster, so that the first important knowledge point cluster is used as the target important knowledge point cluster.

[0116] In some embodiments of the present disclosure, the teaching content evaluation unit 620 is used to: comprehensively obtain a matching verification result of each matching verification according to the first matching result obtained by the following formula 1 and the second matching result obtained by the following formula 2; Formula 1 in, is the first matching result between the knowledge point cluster G and the knowledge point cluster H, where the knowledge point cluster G and the knowledge point cluster H are any two knowledge point clusters that need to be matched and verified among the basic knowledge point cluster, the first important knowledge point cluster, the second important knowledge point cluster, and the third important knowledge point cluster. represents the number of knowledge points in the intersection of knowledge point cluster G and knowledge point cluster H, Represents the number of knowledge points in the union of knowledge point cluster G and knowledge point cluster H; Formula 2 in, is the second matching result between knowledge point cluster G and knowledge point cluster H, n is the nth knowledge point, N is the total number of knowledge points after knowledge point cluster G and knowledge point cluster H are aligned, is the nth knowledge point vector in the knowledge point cluster G, is the mean of the nth knowledge point vector after the knowledge point cluster G and the knowledge point cluster H are aligned, is the nth knowledge point vector in the knowledge point cluster H.

[0117] In some embodiments of the present disclosure, classroom behavior data include interactive behavior data and focused behavior data. The student performance evaluation unit 630 is used to: analyze the students' learning engagement level under classroom teaching based on the interactive behavior data and / or focused behavior data corresponding to the monitoring events triggered by classroom behavior operations under classroom teaching, combined with a preset behavior performance evaluation algorithm, to obtain students' classroom engagement level evaluation indicators. The monitoring events include interactive monitoring events and focused monitoring events.

[0118] In some embodiments of the present disclosure, the data collection unit 610 is used to: in response to interaction monitoring events triggered by classroom behavior operations under classroom teaching, collect teacher interaction audio and student interaction audio corresponding to the interaction monitoring events; identify teacher interaction topics corresponding to the teacher interaction audio and student interaction topics corresponding to the student interaction audio, and determine whether the teacher interaction topic matches the student interaction topic to obtain a matching result; based on the matching result, obtain interaction behavior data, the interaction behavior data including at least one of the interaction duration, the number of interaction failures, and the number of interactions.

[0119] In some embodiments of the present disclosure, the data collection unit 610 is used to: respond to a concentration monitoring event triggered by a classroom behavior operation under classroom teaching, collect student classroom performance data under a classroom mode corresponding to the classroom teaching, the classroom mode is obtained by analyzing classroom instructions corresponding to the classroom behavior operation, the classroom mode includes a discussion mode and a concentration mode, and the student classroom performance data includes student audio and student facial features; based on the student audio and learning facial features, obtain concentration behavior data, the concentration behavior data includes at least one of the effective discussion duration, the invalid discussion duration, and the number of inattentive students.

[0120] In some embodiments of the present disclosure, the data collection unit 610 is used to: collect students' evaluation scores on the classroom teaching content contained in the classroom teaching content; the device also includes: a satisfaction evaluation unit, which is used to evaluate students' satisfaction with the classroom teaching content based on the number of scored students and each student's evaluation score on the classroom teaching content, combined with a preset satisfaction statistical algorithm, so as to evaluate the classroom teaching quality based on the teaching situation, satisfaction level and performance in accordance with a preset teaching quality evaluation algorithm, and obtain a classroom teaching quality evaluation result.

[0121] In some embodiments of the present disclosure, the teaching quality evaluation unit 640 is used to: determine the classroom teaching quality evaluation result based on the teaching goal achievement degree evaluation index of the teaching situation and the student classroom engagement degree evaluation index of the student performance situation, combined with the preset teaching quality evaluation algorithm of the following formula 3; Formula 3 Among them, K is the result of classroom teaching quality evaluation, is the weighted coefficient of the mth teaching situation evaluation index, is the mth teaching situation evaluation indicator.

[0122] It should be noted that the above explanation of the method embodiment is also applicable to the device of this embodiment, and the principle is the same, which is not limited in this embodiment.

[0123] Based on the above Figures 2 to 4 The method shown in the figure, accordingly, this embodiment also provides a computer program product, including a computer program, which implements the above-mentioned Figures 2 to 4 The method shown.

[0124] Based on the above Figures 2 to 4 The method shown in the embodiment also provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the above-mentioned Figures 2 to 4 The method shown.

[0125] Based on this understanding, the technical solution of the present application can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, USB flash drive, mobile hard disk, etc.), including a number of instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute the methods of various implementation scenarios of the present application.

[0126] like Fig. 9 The figure shows a hardware structure diagram of an electronic device of the present invention, including: at least one processor 701; and, A memory 702 that is communicatively connected to at least one of the processors 701; wherein, The memory 702 stores instructions that can be executed by at least one of the processors. The instructions are executed by at least one of the processors so that at least one of the processors can execute the teaching quality assessment method applied to the adaptive teaching quality assessment system as described above.

[0127] Fig. 9 A processor 701 is taken as an example.

[0128] The electronic device may further include: an input device 703 and a display device 704 .

[0129] The processor 701, the memory 702, the input device 703 and the display device 704 may be connected via a bus or other means, and the figure takes the connection via a bus as an example.

[0130] The memory 702 is a non-volatile computer-readable storage medium that can be used to store non-volatile software programs, non-volatile computer executable programs and modules, such as program instructions / modules corresponding to the review content generation method in the embodiment of the present application, for example, Figures 2 to 4 The processor 701 executes various functional applications and data processing by running the non-volatile software programs, instructions and modules stored in the memory 702, that is, the teaching quality evaluation method applied to the adaptive teaching quality evaluation system in the above embodiment is implemented.

[0131] The memory 702 may include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required by at least one function; the data storage area may store data created according to the use of the review content generation method, etc. In addition, the memory 702 may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other non-volatile solid-state storage devices. In some embodiments, the memory 702 may optionally include a memory remotely arranged relative to the processor 701, and these remote memories may be connected to the device for executing the review content generation method via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0132] The input device 703 can receive user clicks and generate signal inputs related to user settings and function controls of the review content generation method. The display device 704 can include display devices such as a display screen.

[0133] The one or more modules are stored in the memory 702, and when executed by the one or more processors 701, the teaching quality assessment method applied to the adaptive teaching quality assessment system in any of the above method embodiments is executed.

[0134] Optionally, the above-mentioned physical device may also include a user interface, a network interface, a camera, a radio frequency (RF) circuit, a sensor, an audio circuit, a WI-FI module, etc. The user interface may include a display, an input unit such as a keyboard, etc., and the optional user interface may also include a USB interface, a card reader interface, etc. The network interface may optionally include a standard wired interface, a wireless interface (such as a WI-FI interface), etc.

[0135] Those skilled in the art will appreciate that the above-mentioned physical device structure provided in this embodiment does not constitute a limitation on the physical device, and may include more or fewer components, or a combination of certain components, or different arrangements of components.

[0136] The storage medium may also include an operating system and a network communication module. The operating system is a program that manages the hardware and software resources of the above-mentioned physical device, and supports the operation of the information processing program and other software and / or programs. The network communication module is used to realize the communication between the components inside the storage medium, and the communication with other hardware and software in the information processing physical device.

[0137] Through the description of the above implementation methods, the technicians in this field can clearly understand that the present application can be implemented by means of software plus necessary general hardware platforms, or by hardware. By applying the scheme of this embodiment, compared with the current existing technology, this embodiment collects the teaching knowledge points contained in the classroom teaching content and the classroom behavior data in the classroom teaching process; the teaching situation of the target important knowledge points in the classroom teaching content is evaluated based on the similarity between the teaching knowledge points and the target important knowledge point clusters, and the target important knowledge point clusters are obtained after extraction from the teaching material content and matching optimization; the students' learning input degree is analyzed based on the classroom behavior data to evaluate the students' performance in the classroom teaching; according to the preset teaching quality evaluation algorithm, the classroom teaching quality is evaluated based on the teaching situation and the performance situation, and the classroom teaching quality evaluation result is obtained, which effectively solves the problems of inaccurate subjective evaluation and one-sided single performance evaluation in the relevant technology, and through objective analysis of the teaching knowledge point coverage and the student input degree, combined with the preset algorithm to comprehensively evaluate the classroom teaching quality, the accuracy and comprehensiveness of the evaluation results are significantly improved, which provides strong support for teaching improvement.

[0138] It should be noted that, in this article, relational terms such as "first" and "second" are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

[0139] The above is only a specific implementation of the present application, so that those skilled in the art can understand or implement the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments described herein, but will conform to the widest scope consistent with the principles and novel features applied for herein.

Claims

1. An adaptive teaching quality evaluation system based on machine learning, characterized in that: The system comprises: Evaluation data collection unit, teacher teaching content evaluation unit, classroom learning evaluation unit and teaching quality evaluation unit; The evaluation data collection unit is used to collect teaching knowledge points contained in classroom teaching content and classroom behavior data during classroom teaching; The teacher teaching content evaluation unit is used to evaluate the teaching situation of the target important knowledge point in the classroom teaching content based on the similarity between the teaching knowledge point and the target important knowledge point cluster, and the target important knowledge point cluster is extracted from the teaching material content and obtained after matching optimization; The classroom learning situation evaluation unit is used to analyze the students' learning engagement based on the classroom behavior data to evaluate the students' performance in classroom teaching; The teaching quality assessment unit is used to assess the classroom teaching quality based on the teaching situation and the performance situation according to a preset teaching quality assessment algorithm to obtain a classroom teaching quality assessment result.

2. The system according to claim 1, characterized in that The teacher teaching content evaluation unit is used to: Extracting basic knowledge point clusters and important knowledge point clusters based on the textbook content, and optimizing the basic knowledge points and the important knowledge point clusters to obtain target important knowledge point clusters; Using a preset similarity evaluation algorithm and combining the same-level matching rules, the similarity evaluation is performed on the knowledge points of each level in the target important knowledge point cluster and the teaching knowledge point cluster of the teaching knowledge point in turn to obtain the knowledge point similarity result; Statistics are collected on teaching objective achievement evaluation indicators corresponding to similar results of the knowledge points.

3. The system according to claim 2, characterized in that The important knowledge point clusters include a first important knowledge point cluster, a second important knowledge point cluster and a third important knowledge point cluster, and the teacher teaching content evaluation unit is used to: Performing hierarchical processing on the textbook text knowledge points corresponding to the textbook content to obtain the basic knowledge point clusters; The first important knowledge point cluster is obtained by screening based on the distribution of knowledge points in the basic knowledge point cluster; Using a preset knowledge point screening model, screening qualified knowledge points from all knowledge points corresponding to the history question bank to obtain the second important knowledge point cluster; Using a hierarchical clustering algorithm and a correlation analysis algorithm to filter out wrong question features under a target feature category from wrong question related features, and taking all knowledge points corresponding to the wrong question features as the third important knowledge point cluster; The first important knowledge point cluster is matched and verified based on the basic knowledge point cluster, the second important knowledge point cluster and the third important knowledge point cluster, and the first important knowledge point cluster is optimized based on the matching verification result to obtain the target important knowledge point cluster.

4. The system according to claim 3, characterized in that The teacher teaching content evaluation unit is used to: Based on the same-level matching rule, using the second important knowledge point cluster and the third important knowledge point cluster, sequentially performing a first matching verification on all the first important knowledge points at each level in the first important knowledge point cluster; If the first matching verification fails, cross-matching verification is performed on each second important knowledge point in the second important knowledge point cluster and each third important knowledge point in the third important knowledge point cluster; If the cross-matching verification is successful, determining the cross-matching knowledge points for which the cross-matching verification is successful, and based on the same-level matching rule, using the cross-matching knowledge points to sequentially perform a second matching verification on all basic knowledge points at each level in the basic knowledge point cluster; If the second matching verification is successful, the matching knowledge points corresponding to the second matching verification success are integrated into the first important knowledge point cluster, so that the first important knowledge point cluster is used as the target important knowledge point cluster.

5. The system according to claim 3, characterized in that The teacher teaching content evaluation unit is used to: The first matching result obtained by the following formula 1 and the second matching result obtained by the following formula 2 are used to obtain the matching verification result of each matching verification; Formula 1 in, is the first matching result between the knowledge point cluster G and the knowledge point cluster H, wherein the knowledge point cluster G and the knowledge point cluster H are any two knowledge point clusters that need to be matched and verified among the basic knowledge point cluster, the first important knowledge point cluster, the second important knowledge point cluster, and the third important knowledge point cluster, represents the number of knowledge points in the intersection of knowledge point cluster G and knowledge point cluster H, Represents the number of knowledge points in the union of knowledge point cluster G and knowledge point cluster H; Formula 2 in, is the second matching result between the knowledge point cluster G and the knowledge point cluster H, n is the nth knowledge point, N is the total number of knowledge points after the knowledge point cluster G and the knowledge point cluster H are aligned, is the nth knowledge point vector in the knowledge point cluster G, is the mean value of the nth knowledge point vector after the knowledge point cluster G and the knowledge point cluster H are aligned, is the nth knowledge point vector in the knowledge point cluster H.

6. The system according to claim 1, characterized in that The classroom behavior data includes interactive behavior data and focused behavior data, and the classroom learning situation evaluation unit is used to: Based on the interactive behavior data and / or focused behavior data corresponding to the monitoring events triggered by the classroom behavior operations under the classroom teaching, combined with a preset behavior performance evaluation algorithm, the students' learning engagement level under the classroom teaching is analyzed to obtain an evaluation index of the students' classroom engagement level. The monitoring events include interactive monitoring events and focused monitoring events.

7. The system according to claim 6, characterized in that The evaluation data collection unit is used to: In response to an interactive monitoring event triggered by a classroom behavior operation under the classroom teaching, collecting teacher interactive audio and student interactive audio corresponding to the interactive monitoring event; Identify the teacher interaction topic corresponding to the teacher interaction audio and the student interaction topic corresponding to the student interaction audio, and determine whether the teacher interaction topic matches the student interaction topic to obtain a matching result; Based on the matching result, the interactive behavior data is obtained, and the interactive behavior data includes at least one of the interaction duration, the number of interaction failures, and the number of interactions.

8. The system according to claim 6, characterized in that The evaluation data collection unit is used to: In response to a concentration monitoring event triggered by a classroom behavior operation under the classroom teaching, collecting student classroom performance data in a classroom mode corresponding to the classroom teaching, wherein the classroom mode is obtained by analyzing classroom instructions corresponding to the classroom behavior operation, the classroom mode includes a discussion mode and a concentration mode, and the student classroom performance data includes student audio and student facial features; Based on the student audio and learning facial features, the concentration behavior data is obtained, and the concentration behavior data includes at least one of the effective discussion duration, the invalid discussion duration and the number of inattentive students.

9. The system according to claim 1, characterized in that The system also includes: a student teaching experience evaluation unit, which is used to obtain the student evaluation scores of the classroom teaching content contained in the classroom teaching content collected by the evaluation data collection unit; based on the number of scored students and each student's evaluation score of the classroom teaching content, combined with a preset satisfaction statistical algorithm, evaluate the student's satisfaction with the classroom teaching content, so as to evaluate the classroom teaching quality based on the teaching situation, the satisfaction level and the performance according to the preset teaching quality evaluation algorithm, and obtain a classroom teaching quality evaluation result.

10. The system according to claim 1, characterized in that The teaching quality assessment unit is used to: Based on the teaching goal achievement evaluation index of the teaching situation and the student classroom engagement evaluation index of the student performance situation, combined with the preset teaching quality evaluation algorithm of the following formula 3, determine the classroom teaching quality evaluation result; Formula 3 Wherein, K is the classroom teaching quality evaluation result, is the weighted coefficient of the mth evaluation index, is the mth evaluation indicator.

Images