An English teaching quality evaluation method based on a knowledge graph

Abstract

The present invention relates to the field of English teaching, and particularly to an English teaching quality evaluation method based on a knowledge graph. The technical problem to be solved by the present invention is to provide an English teaching quality evaluation method based on a knowledge graph that can combine multi-dimensional information to comprehensively and objectively evaluate the quality of English teaching. An English teaching quality evaluation method based on a knowledge graph achieves the effect of combining multi-dimensional information of students' learning data, teaching data of teachers, and teaching resource information to comprehensively and objectively evaluate the quality of English teaching.

Description

Technical Field

[0001] The present invention relates to the field of English teaching, and particularly to an English teaching quality evaluation method based on a knowledge graph. Background Art

[0002] In recent years, with the wide application of information technology and digital technology in the field of education, our learning methods have undergone great changes, breaking the limitations of traditional teaching activities in terms of time and space. As a result, online learning has received extensive attention. Online learning mainly refers to the way of learning through the Internet, and students can obtain resources such as texts, videos, and pictures related to course knowledge from the Internet according to their actual needs. Such an open learning method has lowered the threshold of knowledge acquisition, improved the utilization ability of online education resources, and greatly promoted educational equity, which has extremely profound significance.

[0003] With the development of Internet technology, various teaching management systems have gradually emerged, which have greatly facilitated the teaching management work of schools. However, the teaching quality evaluation system in the existing teaching management systems still uses traditional evaluation methods, which have problems such as large workload, small sampling range, untrue evaluation results, and high contingency, reducing the actual value of the evaluation. Therefore, there is an urgent need to develop an English teaching quality evaluation method based on a knowledge graph that can combine multi-dimensional information to comprehensively and objectively evaluate the English teaching quality. Summary of the Invention

[0004] In order to overcome the problems of large workload, small sampling range, untrue evaluation results, and high contingency in the teaching quality evaluation system of the existing teaching management system, which reduce the actual value of the evaluation, the technical problem to be solved by the present invention is to provide an English teaching quality evaluation method based on a knowledge graph that can combine multi-dimensional information to comprehensively and objectively evaluate the English teaching quality.

[0005] The present invention is achieved by the following specific technical means:

[0006] An English teaching quality evaluation method based on a knowledge graph, comprising the following steps:

[0007] a. Multi-source data collection and processing:

[0008] Collect structured data, including student terminal learning data, classroom attendance records, and teacher evaluation scores;

[0009] Collect unstructured data, including classroom recordings, teaching videos, student homework texts, and teacher lesson plans;

[0010] Collect teaching environment data, and obtain classroom acoustic parameters, light intensity, seat distribution density, and teaching equipment usage status through Internet of Things devices;

[0011] Clean, deduplicate, and standardize multi-source data, and perform semantic annotation to generate a unified teaching dataset;

[0012] b. Construction of teaching knowledge graph:

[0013] Construct a knowledge graph ontology framework including a teaching subject layer, a knowledge system layer, a teaching process layer, and an effect evaluation layer; the teaching subject layer includes teacher nodes with teaching experience attributes and student nodes with learning style attributes, the knowledge system layer includes a grammar knowledge point network and a lexical semantic association matrix, the teaching process layer includes teaching activity nodes annotated with spatio-temporal features, and the effect evaluation layer includes the mapping relationship between formative evaluation indicators and summative evaluation, where the formative evaluation indicators include the frequency of classroom interaction and the summative evaluation includes test scores;

[0014] Use the BERT-BiLSTM-CRF model to process unstructured text, and combine the TransE algorithm for vector representation of entity relationships;

[0015] Complete missing triples through knowledge reasoning. When the average fixation duration of a certain knowledge point in the newly added teaching video analysis exceeds the threshold, automatically enhance the weight of this node;

[0016] c. Construction of intelligent evaluation model:

[0017] Extract quantitative evaluation indicators and qualitative evaluation indicators from the knowledge graph; among them, the quantitative evaluation indicators include knowledge point coverage rate, teaching resource turnover rate, and online learning activity, and the qualitative evaluation indicators include teaching objective achievement degree and classroom atmosphere index;

[0018] Construct a hybrid evaluation model, use a graph neural network to analyze the association relationship between teachers-students-knowledge points, and dynamically adjust the weights of each indicator through an attention mechanism;

[0019] Use a time series model to track the long-term changes in teaching effects;

[0020] Integrate an expert rule base, the expert rule base includes a teaching effect attenuation rule and a learning path optimization rule, where the teaching effect attenuation rule is to trigger curriculum revision when the error rate exceeds the threshold for 3 consecutive evaluations, and the learning path optimization rule is to insert intensive training when a weak association of "adjective usage → writing application" is detected;

[0021] d. Visualization and optimization feedback:

[0022] Generate a three-dimensional visualization map, where the node size represents the importance of elements, the edge weight shows the association strength, and the color coding indicates the compliance status;

[0023] Synchronously output the teacher ability radar chart, knowledge point heat map, and improvement path decision tree;

[0024] Compare the evaluation results with the actual improvement effects through the reinforcement learning algorithm, and dynamically update the graph relationship weights;

[0025] Push personalized strategies, including recommending a teaching case library to the teacher side and prompting to adjust the interaction frequency, and generating an adaptive path containing remedial exercises and micro-lesson videos for the student side;

[0026] Build virtual teaching nodes, simulate the implementation effects of strategies such as flipped classrooms, and generate an optimal plan prediction report.

[0027] Furthermore, the student terminal learning data described in step a includes the number of times of repeated learning of knowledge points and the standard deviation of test score fluctuations; the teaching environment data includes the regional speech clarity differences obtained through distributed acoustic sensors.

[0028] Furthermore, the construction of the knowledge system layer in step b includes:

[0029] Explicit knowledge network: Build a predecessor-successor relationship graph of grammar rules based on curriculum standards;

[0030] Implicit knowledge network: Establish a pragmatic strategy association matrix for cross-cultural communication scenarios;

[0031] Dynamic update mechanism: When the correct rate of the associated test of a certain knowledge point continues to decline, automatically trigger the weight attenuation of the teaching resources connected to this node.

[0032] Furthermore, the expert rule library in step c also includes:

[0033] Teaching strategy verification rule: If the improvement of the goal achievement degree after adopting the task-driven method fails to meet the expectation, then reduce the method adaptation degree score;

[0034] Environment adaptation rule: When the noise decibel value continues to be higher than 55 dB, trigger teaching venue adjustment suggestions.

[0035] Furthermore, the three-dimensional visualization graph in step d supports:

[0036] Multi-level drill-down analysis: Drill down from the school-wide curriculum network to the knowledge point mastery path of a single student;

[0037] Causal path tracing: Associate classroom interaction records with later test performances, and generate a teaching intervention effectiveness report;

[0038] Virtual scenario comparison: Side by side display the index differences between actual teaching and virtual strategies.

[0039] Furthermore, it also includes building a teaching effect prediction model and generating an optimal strategy through the following steps:

[0040] Inject virtual teaching behavior nodes to construct parallel teaching scenarios;

[0041] Use a temporal graph convolutional network to predict the trend of index changes in each scenario;

[0042] Calculate the risk - return ratio of the strategy implementation and output an improvement plan including priority ranking.

[0043] Furthermore, an English teaching quality assessment system includes:

[0044] Multi - source data acquisition module: Integrate Internet of Things sensors, learning platform APIs, and audio - video analysis units;

[0045] Knowledge graph engine: A distributed graph database that supports dynamic ontology modeling and relationship reasoning;

[0046] Hybrid computing core: Includes a GNN - TCN joint model and a Drools rule engine;

[0047] Adaptive feedback module: A strategy optimizer with reinforcement learning capabilities;

[0048] Multi - terminal interaction platform: A visualization system that supports teachers, students, and administrators to access according to permissions.

[0049] Furthermore, a computer - readable storage medium stores a computer program, and when the program is executed by a processor, it implements the steps of the method described in any one of the above.

[0050] Compared with the prior art, the present invention has the following beneficial effects:

[0051] The present invention achieves the effect of comprehensively and objectively evaluating the English teaching quality by combining multi - dimensional information of students' learning data, teachers' teaching data, and teaching resource information. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 It is a schematic diagram of the steps of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0053] The following further describes the present invention with reference to the accompanying drawings: Embodiment

[0054] An English teaching quality assessment method based on a knowledge graph includes the following steps:

[0055] a. Multi - source data acquisition and processing:

[0056] Collect structured data, including students' terminal learning data, classroom attendance records, and teachers' evaluation scores;

[0057] Collect unstructured data, including classroom recordings, teaching videos, students' assignment texts, and teachers' lesson plans;

[0058] Collect teaching environment data, and obtain classroom acoustic parameters, light intensity, seat distribution density, and the usage status of teaching equipment through Internet of Things devices;

[0059] Clean, deduplicate, and standardize multi-source data, and perform semantic annotation to generate a unified teaching dataset;

[0060] b. Construction of teaching knowledge graph:

[0061] Construct a knowledge graph ontology framework including a teaching subject layer, a knowledge system layer, a teaching process layer, and an effect evaluation layer; the teaching subject layer includes teacher nodes with teaching experience attributes and student nodes with learning style attributes, the knowledge system layer includes a grammar knowledge point network and a lexical semantic association matrix, the teaching process layer includes teaching activity nodes marked with spatio-temporal features, and the effect evaluation layer includes the mapping relationship between formative evaluation indicators and summative evaluation, where the formative evaluation indicators include the frequency of classroom interaction, and the summative evaluation includes test scores;

[0062] Use the BERT-BiLSTM-CRF model to process unstructured text, and combine the TransE algorithm for vector representation of entity relationships;

[0063] Complete missing triples through knowledge reasoning. When the average fixation duration of a certain knowledge point in the newly added teaching video analysis exceeds the threshold, automatically enhance the weight of this node;

[0064] c. Construction of intelligent evaluation model:

[0065] Extract quantitative evaluation indicators and qualitative evaluation indicators from the knowledge graph; among them, the quantitative evaluation indicators include knowledge point coverage rate, teaching resource turnover rate, and online learning activity, and the qualitative evaluation indicators include teaching goal achievement degree and classroom atmosphere index;

[0066] Construct a hybrid evaluation model, use a graph neural network to analyze the association relationship between teachers-students-knowledge points, and dynamically adjust the weights of each indicator through an attention mechanism;

[0067] Use a time series model to track the long-term changes in teaching effects;

[0068] Integrate an expert rule base, and the expert rule base includes a teaching effect attenuation rule and a learning path optimization rule. Among them, the teaching effect attenuation rule is to trigger curriculum revision when the error rate of continuous 3 evaluations exceeds the threshold, and the learning path optimization rule is to insert intensive training when a weak association of "adjective usage → writing application" is detected;

[0069] d. Visualization and optimization feedback:

[0070] Generate a 3D visualization graph, where the node size represents the importance of elements, the edge weight shows the association strength, and the color coding indicates the compliance status;

[0071] Synchronously output the teacher ability radar chart, knowledge point heat map, and improvement path decision tree;

[0072] Compare the evaluation results with the actual improvement effects through the reinforcement learning algorithm, and dynamically update the graph relationship weights;

[0073] Push personalized strategies, including recommending a teaching case library to the teacher side and prompting to adjust the interaction frequency, and generating an adaptive path containing remedial exercises and micro-lesson videos for the student side;

[0074] Construct virtual teaching nodes, simulate the implementation effects of strategies such as flipped classrooms, and generate an optimal solution prediction report.

[0075] Working principle:

[0076] This method is based on knowledge graph technology and realizes the intelligent evaluation and optimization of English teaching quality through multi-source data fusion and dynamic modeling. Its core principle is as follows:

[0077] First, through the integration of multi-source heterogeneous data, the system collects students' terminal learning behavior data (such as the frequency of repeated learning of knowledge points and the fluctuation of test scores), teachers' teaching process data (classroom recordings, teaching plan texts), and teaching environment parameters (acoustics, lighting, seat density), and conducts standardized cleaning and semantic annotation on them to construct a unified teaching data set. On this basis, a multi-layer teaching knowledge graph is constructed: the teaching subject layer depicts the attribute associations between teachers and students, the knowledge system layer integrates explicit grammar rules and implicit pragmatic strategies, the teaching process layer records teaching activities in the space-time dimension, and the effect evaluation layer establishes a dynamic mapping relationship between formative evaluation (such as the frequency of classroom interaction) and summative evaluation (such as test scores). Use the BERT-BiLSTM-CRF model to extract the semantic features of unstructured text, combine the TransE algorithm for vector representation of entity relationships, and automatically complete missing associations through knowledge reasoning (such as dynamically adjusting the knowledge point weights according to the student's video gaze duration).

[0078] Secondly, construct a hybrid intelligent evaluation model: based on the Graph Neural Network (GNN) to mine the potential patterns in the teacher-student-knowledge point triadic relationship, use the Temporal Convolutional Network (TCN) to track the long-term evolution trend of teaching effects, and dynamically allocate the weights of evaluation indicators through the attention mechanism (such as the combined influence of the classroom atmosphere index and the knowledge point coverage rate). Integrate the expert rule base to achieve closed-loop optimization: when detecting a weak association between "adjective usage → writing application", automatically insert a special training module; if the environmental noise continues to exceed the standard, trigger suggestions for adjusting the teaching venue. At the same time, the reinforcement learning-driven feedback system dynamically updates the weights of the knowledge graph relationships by comparing the simulation effects of virtual teaching strategies (such as flipped classrooms) with the actual improvement data, generates a three-dimensional visualization graph to reveal the association strength of teaching elements, and pushes a personalized case library to teachers and recommends an adaptive learning path to students, forming a continuous improvement mechanism of "evaluation - feedback - optimization".

[0079] Through the deep integration of data-driven and knowledge reasoning, this system realizes the full-chain evaluation from the mastery of micro knowledge points to the optimization of macro teaching strategies, significantly improving the accuracy and intervention timeliness of English teaching quality analysis.

[0080] Although the present disclosure has been described in detail with reference to exemplary embodiments, the present disclosure is not limited thereto, and it will be apparent to those skilled in the art that various modifications and changes can be made thereto without departing from the scope of the present disclosure.

Claims

1. An English teaching quality evaluation method based on a knowledge graph, characterized in that The method includes the following steps: a. Multi-source data collection and processing: Collect structured data, including student terminal learning data, classroom attendance records, and teacher evaluation scores; Collect unstructured data, including classroom recordings, teaching videos, student assignment texts, and teacher lesson plans; Collect teaching environment data, and obtain classroom acoustic parameters, light intensity, seat distribution density, and teaching equipment usage status through Internet of Things devices; Clean, deduplicate, and standardize the multi-source data, and perform semantic annotation to generate a unified teaching dataset; b. Construction of teaching knowledge graph: Construct a knowledge graph ontology framework including a teaching subject layer, a knowledge system layer, a teaching process layer, and an effect evaluation layer; the teaching subject layer includes teacher nodes with teaching experience attributes and student nodes with learning style attributes, the knowledge system layer includes a grammar knowledge point network and a lexical semantic association matrix, the teaching process layer includes teaching activity nodes marked with spatio-temporal characteristics, and the effect evaluation layer includes the mapping relationship between formative evaluation indicators and summative evaluation, where the formative evaluation indicators include the frequency of classroom interaction, and the summative evaluation includes test scores; Use the BERT-BiLSTM-CRF model to process unstructured text, and combine the TransE algorithm for vector representation of entity relationships; Complete the missing triples through knowledge reasoning. When the analysis of newly added teaching videos shows that the average fixation duration of a certain knowledge point exceeds the threshold, automatically enhance the weight of this node; c. Construction of intelligent evaluation model: Extract quantitative evaluation indicators and qualitative evaluation indicators from the knowledge graph; among them, the quantitative evaluation indicators include knowledge point coverage rate, teaching resource turnover rate, and online learning activity, and the qualitative evaluation indicators include teaching goal achievement degree and classroom atmosphere index; Construct a hybrid evaluation model, use a graph neural network to analyze the association relationship between teachers-students-knowledge points, and dynamically adjust the weights of each indicator through an attention mechanism; Use a time series model to track the long-term changes in teaching effects; Integrate an expert rule base, which includes a teaching effect attenuation rule and a learning path optimization rule. The teaching effect attenuation rule is to trigger curriculum revision when the evaluation error rate exceeds the threshold for 3 consecutive times, and the learning path optimization rule is to insert intensive training when a weak association of "adjective usage → writing application" is detected; d. Visualization and optimization feedback: Generate a three-dimensional visualization graph, where the node size represents the importance of elements, the edge weight shows the association strength, and the color coding indicates the compliance status; Synchronously output a teacher ability radar chart, a knowledge point heat map, and an improvement path decision tree; Compare the evaluation results with the actual improvement effects through a reinforcement learning algorithm, and dynamically update the graph relationship weights; Push personalized strategies, where recommend a teaching case library to the teacher side and prompt to adjust the interaction frequency, and generate an adaptive path including remedial exercises and micro-lesson videos for the student side; Construct a virtual teaching node, simulate the implementation effects of strategies such as flipped classrooms, and generate an optimal solution prediction report.

2. The method for evaluating the quality of English teaching based on a knowledge graph according to claim 1, wherein The student terminal learning data described in step a includes the number of times of repeated learning of knowledge points and the standard deviation of test score fluctuations; the teaching environment data includes the regional speech clarity difference obtained through distributed acoustic sensors.

3. The method for evaluating the quality of English teaching based on a knowledge graph according to claim 1, characterized in that, The construction of the knowledge system layer in step b includes: Explicit knowledge network, constructing a predecessor-successor relationship graph of grammar rules based on curriculum standards; Implicit knowledge network, establishing a pragmatic strategy association matrix for cross-cultural communication scenarios; Dynamic update mechanism, when the correct rate of the correlation test of a certain knowledge point continuously decreases, automatically trigger the weight attenuation of the teaching resources connected to this node.

4. A method for evaluating the quality of English teaching based on a knowledge graph according to claim 1, characterized in that, The expert rule base in step c also includes: Teaching strategy verification rule: If the improvement of the goal achievement degree after adopting the task-driven method does not meet the expectation, then reduce the method adaptation degree score; Environment adaptation rule: When the noise decibel value continuously exceeds 55 dB, trigger suggestions for adjusting the teaching place.

5. A method for evaluating the quality of English teaching based on a knowledge graph according to claim 1, characterized in that, The three-dimensional visualization map in step d supports: Multi-level drill-down analysis: Drill down from the school-wide curriculum network to the knowledge point mastery path of a single student; Causal path tracing: Correlate classroom interaction records with later test performance to generate a teaching intervention effectiveness report; Virtual scenario comparison: Display the index differences between actual teaching and virtual strategies side by side.

6. The method for evaluating the quality of English teaching based on a knowledge graph according to claim 1, wherein It also includes constructing a teaching effect prediction model, and generating strategies through the following steps: Inject virtual teaching behavior nodes to construct a parallel teaching scenario; Adopt a temporal graph convolutional network to predict the index change trend in each scenario; Calculate the risk-benefit ratio of strategy implementation, and output an improvement plan including priority ranking.

7. An English teaching quality evaluation system, characterized in that, Including: Multi-source data acquisition module: Integrate Internet of Things sensors, learning platform APIs, and audio-video analysis units; Knowledge graph engine: A distributed graph database that supports dynamic ontology modeling and relationship reasoning; Hybrid computing core: Includes a GNN-TCN joint model and a Drools rule engine; Adaptive feedback module: A strategy optimizer with reinforcement learning capabilities; Multi-terminal interaction platform: A visualization system that supports teachers, students, and administrators to access according to permissions.

8. A computer-readable storage medium storing a computer program, characterized in that, When the program is executed by a processor, it implements the steps of the method described in any one of claims 1-6.

Images