A teaching information processing system and method based on electrical automation control

Abstract

The application discloses a kind of teaching information processing system and method based on electrical automation control, it is related to intelligent teaching technical field, including, acquisition multimodal teaching information, and pretreatment is carried out;Using the multimodal teaching information after pretreatment, extract and fuse feature, generate learning state feature;Learning state feature is used to build and dynamically adjust knowledge graph, obtain dynamic knowledge graph;Based on dynamic knowledge graph generation individualized learning path, and recommend learning resource, record learning resource utilization rate;Based on learning state feature and learning resource utilization rate, generate the comprehensive learning state of student;According to comprehensive learning state generation teaching feedback.The application is by building dynamic knowledge graph, in combination with the learning state feature of student and learning resource utilization rate, dynamically assesses the knowledge point mastering condition of student, generates individualized learning path and pushes the learning resource of adaptation.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of intelligent teaching, and particularly relates to a teaching information processing system and method based on electrical automation control. BACKGROUND

[0002] In the teaching process of electrical automation control courses, traditional teaching methods often rely on the experience of teachers and self-feedback of students, lacking comprehensive and real-time monitoring of students' learning state. With the development of information technology, although some education platforms have begun to introduce data analysis to assist teaching, the existing solutions still have many deficiencies.

[0003] Most existing teaching systems can only process single-modal data such as text or operation records, and cannot comprehensively analyze multi-modal information. This limitation makes it difficult for the system to accurately capture the real learning state of students, thereby affecting the effectiveness of personalized teaching paths. For example, in courses such as electrical automation control which are highly practical, simply relying on homework completion or test scores cannot fully reflect the actual mastery level and problems encountered by students. SUMMARY

[0004] In view of the above existing problems, the present application is proposed.

[0005] Therefore, the present application provides a teaching information processing method based on electrical automation control to solve the problems of insufficient multi-modal data comprehensive analysis and lack of real-time feedback.

[0006] To solve the above technical problems, the present application provides the following technical solutions:

[0007] In a first aspect, the present application provides a teaching information processing method based on electrical automation control, which includes collecting multi-modal teaching information and preprocessing it.

[0008] Using the preprocessed multi-modal teaching information, features are extracted and fused to generate learning state features.

[0009] Using the learning state features, a knowledge graph is constructed and dynamically adjusted to obtain a dynamic knowledge graph.

[0010] Based on the dynamic knowledge graph, a personalized learning path is generated, learning resources are recommended, and learning resource utilization rate is recorded.

[0011] Based on the learning state features and learning resource utilization rate, a comprehensive learning state of the student is generated.

[0012] According to the comprehensive learning state, teaching feedback is generated.

[0013] As a preferred scheme of the teaching information processing method based on electrical automation control, the multi-modal teaching information includes operation data, measurement data and analysis results in experimental and simulation operations of students in the electrical automation control course, experimental report content, homework completion, answer record, learning notes, learning time and learning resource access frequency.

[0014] The preprocessing includes data cleaning, denoising, formatting and spatio-temporal alignment.

[0015] As a preferred scheme of the teaching information processing method based on electrical automation control, the multi-modal teaching information includes operation data, measurement data and analysis results in experimental and simulation operations of students in the electrical automation control course, experimental report content, homework completion, answer record, learning notes, learning time and learning resource access frequency.

[0016] The operation data, measurement data and analysis results in the experimental and simulation operations are extracted and fused by a convolutional neural network to obtain behavior features.

[0017] The experimental report, homework and learning notes are analyzed by a BERT natural language processing method to extract text features.

[0018] The behavior features and text features are fused by a multi-modal attention mechanism, and attention weights need to be calculated before fusion, and the calculation method is represented as

[0019]

[0020] Wherein, α b→t is the attention weight of the behavior feature to the text feature, softmax is a function for converting an input vector into a probability distribution, F' b is the behavior feature, F' t is the text feature, W Q is a pre-linear transformation matrix, W K is a post-linear transformation matrix, α t→b is the attention weight of the text feature to the behavior feature, is a standardization factor.

[0021] The behavior features and text features are fused by a multi-modal attention mechanism to obtain the learning state features of the students, which are represented as

[0022] F s = ReLU (α b→t F' t + α t→b F' b ).

[0023] Wherein, F s is the learning state feature of the student, ReLU is a linear rectification function, αb→t is the attention weight of the behavior feature on the text feature, F t is the text feature, a t→b is the attention weight of the text feature on the behavior feature, F b is the behavior feature.

[0024] As a preferred scheme of the teaching information processing method based on electrical automation control, the knowledge graph is constructed and dynamically adjusted by using the learning state feature, and a dynamic knowledge graph is obtained, and the specific steps are as follows,

[0025] According to the structure of the knowledge points of the electrical automation control course, a knowledge graph with knowledge point nodes, node weights and association relationships between nodes is constructed;

[0026] The learning state feature is used to analyze the mastery of students on each knowledge point, and the weights of the knowledge point nodes in the knowledge graph are initialized;

[0027] According to the learning state feature of the student, the weights of each knowledge point node in the knowledge graph are adjusted in real time, which is expressed as,

[0028]

[0029] Wherein, m i (t) is the weight of the i-th knowledge point node at time t, m i (t-1) is the weight of the i-th knowledge point node at time t-1, e is the base of natural logarithm, η is the decay rate parameter, t is the time, λ is the proportion factor of weight update, ReLU is the linear rectifier function, is the learning state feature vector at time t, and is the weight vector related to the i-th knowledge point;

[0030] According to the mastery of the knowledge points, the association strength between each knowledge point node is dynamically adjusted, which is expressed as,

[0031]

[0032] Wherein, is the association strength between the knowledge point node i and the knowledge point node j at time t, is the initial association strength, tanh is the hyperbolic tangent function, Δm i is the weight change of the i-th knowledge point node, Δm j is the weight change of the j-th knowledge point node, d ij is the distance between the knowledge point node i and the knowledge point node j at time t, and β is the adjustment parameter;

[0033] The adjusted weights and association relationships of the knowledge point nodes are integrated to generate a dynamic knowledge graph that reflects the learning state of the student in real time.

[0034] As a preferred scheme of the teaching information processing method based on electrical automation control, the personalized learning path is generated based on the dynamic knowledge graph, and learning resources are recommended, and the utilization rate of learning resources is recorded, and the specific steps are as follows,

[0035] Referring to the node association relationship in the knowledge graph, the knowledge points associated with the low weight nodes are located as review content by using the adjacency matrix traversal algorithm, and the conditions for generating the learning path planning are generated;

[0036] The personalized learning path is generated by using the Dijkstra shortest path algorithm and according to the learning path planning conditions;

[0037] The learning needs of each knowledge point in the personalized learning path are analyzed, and the learning resources are matched and adapted from the teaching resource library according to the learning needs by using the content-based collaborative filtering recommendation algorithm;

[0038] The access frequency, learning duration and completion degree of the recommended resources by the students are recorded and analyzed, and the resource utilization rate is obtained.

[0039] As a preferred scheme of the teaching information processing method based on electrical automation control, the teaching feedback is generated according to the comprehensive learning state, and the specific steps are as follows,

[0040] According to the comprehensive characteristics of the knowledge point mastery degree and the resource utilization rate, the learning effect of the students is calculated by using the weight distribution algorithm;

[0041] The contribution degree of different knowledge points to the overall learning effect is analyzed by using the association relationship between the knowledge point nodes in the dynamic knowledge graph;

[0042] The priority of the key knowledge points is determined by the structure of the knowledge graph, and the mastery of the students on the knowledge points is summarized to generate the overall learning effect of the students according to the weight data of the extracted knowledge point nodes and the priority of the knowledge points;

[0043] According to the distribution of the knowledge point mastery degree in the comprehensive learning state, the knowledge points not mastered and poorly mastered by the students are identified;

[0044] Based on the identified knowledge points not mastered and poorly mastered by the students, learning resources are pushed by using the dynamic recommendation algorithm;

[0045] After the pushing is completed, the learning data is re-collected and fed back, and the dynamic knowledge graph and the personalized learning path are updated.

[0046] In a second aspect, the present application provides a teaching information processing system based on electrical automation control, comprising an information collection module, a feature extraction module, a knowledge graph module, a learning resource module, a learning state module and a feedback module.

[0047] The information collection module is configured to collect multi-modal teaching information and perform preprocessing.

[0048] The feature extraction module is configured to extract and fuse features using the preprocessed multi-modal teaching information to generate learning state features.

[0049] The knowledge graph module is configured to construct and dynamically adjust a knowledge graph using the learning state features to obtain a dynamic knowledge graph.

[0050] The learning resource module is configured to generate a personalized learning path based on the dynamic knowledge graph, recommend learning resources and record learning resource utilization.

[0051] The learning state module is configured to generate a comprehensive learning state of a student based on the learning state features and the learning resource utilization.

[0052] The feedback optimization module is configured to generate teaching feedback according to the comprehensive learning state.

[0053] In a third aspect, the present application provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, and wherein the computer program, when executed by the processor, implements any step of the teaching information processing method based on electrical automation control according to the first aspect of the present application.

[0054] In a fourth aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements any step of the teaching information processing method based on electrical automation control according to the first aspect of the present application.

[0055] The present application has the following beneficial effects: The present application dynamically assesses the knowledge point mastery of students by constructing a dynamic knowledge graph, combining the learning state features and learning resource utilization of students, generating a personalized learning path and pushing adaptive learning resources. The present application can update the weights and association relationships of each knowledge point node in the knowledge graph in real time, reflect the learning progress and weak links of students, thereby dynamically adjusting the teaching strategy to improve the pertinence and individualization of teaching. The present application constructs a student-specific learning path through a dynamic knowledge graph, optimizes the learning efficiency, provides a real-time feedback mechanism, helps students quickly find weak points in learning, and improves the mastery of knowledge points through accurate recommendation of adaptive resources. The present application realizes closed-loop management of learning data, continuously optimizes teaching resources and path planning through collection, analysis and feedback of student learning data. BRIEF DESCRIPTION OF DRAWINGS

[0056] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0057] Figure 1 This is a flow chart of the teaching information processing method based on electrical automation control in Example 1.

[0058] Figure 2 This is a module diagram of the teaching information processing system based on electrical automation control in Example 1. DETAILED DESCRIPTION

[0059] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0060] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0061] Secondly, the term "one embodiment" or "embodiment" herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in various places throughout this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive of other embodiments.

[0062] Example 1, reference Figure 1 and Figure 2 , which is the first embodiment of the present invention, provides a teaching information processing method based on electrical automation control, comprising the following steps:

[0063] S1. Collect multimodal teaching information and perform preprocessing, including the following steps:

[0064] The multi-modal teaching information includes operation data, measurement data and analysis results of students in experiment and simulation operation of the electrical automation control course, experiment report content, homework completion, answer record, learning notes, learning time and learning resource access frequency. Specifically, the operation sequence, operation parameters and measurement data of students in the experiment and simulation operation experiment process of the electrical automation control course are collected in real time. The report text and homework answers submitted by the students. The answer record is the answer data of the students in the online practice and examination; the learning resource access record is the frequency and duration of the students accessing the learning resources (such as electronic textbooks, video explanation); the learning notes are the note contents recorded by the students in the learning process.

[0065] The preprocessing includes data cleaning, denoising, formatting and space-time alignment. Specifically, incomplete, repeated or invalid records are removed, such as missing steps or abnormal values in experiment operation; the noise in the measurement data is smoothed by using the median filtering method; the experiment operation data, resource access record and the like are aligned according to the time stamp, so as to ensure that different modal information is analyzed on the same time axis.

[0066] It should be noted that the adopted preprocessing method can effectively improve the quality of the multi-modal data, ensure the consistency of different types of data in time and space, and thus lay a foundation for subsequent feature extraction and fusion. Through comprehensive collection and standardized processing of multi-modal teaching information, the present application can accurately capture the learning behavior and state of students, and provide high-quality data support for the construction of dynamic knowledge graph.

[0067] S2, using the preprocessed multi-modal teaching information, extracting and fusing features to generate learning state features, including the following steps,

[0068] The operation data, measurement data and analysis results in the experiment and simulation operation are extracted by using the convolutional neural network to obtain the behavior features. Specifically, the operation sequence is segmented, and the key operation features (such as operation time, parameter change trend) of each segment are extracted; the measurement data is locally pattern captured, and the operation precision and consistency features of the students in the experiment are extracted; the analysis results are classified, and whether the students complete the correct conclusion in the experiment is identified.

[0069] The text features are extracted by using the BERT natural language processing method (bidirectional encoder representation model) to analyze the experiment report, homework and learning notes. Specifically, the text is segmented and input into the BERT model to extract semantic embedding representation; the semantic features of the key sentences in the student report are extracted, such as experiment purpose, method and conclusion.

[0070] The behavior features and text features are fused by using the multi-modal attention mechanism to fuse the behavior features and text features. The attention weight needs to be calculated before fusion, and the calculation method is represented as,

[0071]

[0072]

[0073] wherein, alpha b→t is the attention weight of the behavior feature to the text feature, softmax is a function for converting an input vector into a probability distribution, F' b is the behavior feature, F' t is the text feature, W Q is a pre-linear transformation matrix, W K is a post-linear transformation matrix, alpha t→b is the attention weight of the text feature to the behavior feature, is a standardization factor.

[0074] The behavior feature and the text feature are fused through the multi-modal attention mechanism to obtain the learning state feature of the student, which is represented as,

[0075] F s = ReLU(alpha b→t F' t + alpha t→b F' b );

[0076] wherein, F s is the learning state feature of the student, ReLU is a linear rectifier function, alpha b→t is the attention weight of the behavior feature to the text feature, F' t is the text feature, alpha t→b is the attention weight of the text feature to the behavior feature, F' b is the behavior feature.

[0077] It should be noted that the behavior feature is extracted by using the convolutional neural network, which can effectively capture the behavior pattern of the student in the experiment and simulation operation; the text feature is extracted by using BERT, which can deeply mine the semantic information in the learning text of the student; the feature fusion is realized through the multi-modal attention mechanism, which can comprehensively reflect the learning state of the student. By fusing the behavior feature and the text feature, the learning state of the student can be more comprehensively and accurately described, and a reliable basis can be provided for the initialization and updating of the dynamic knowledge graph.

[0078] S3, constructing and dynamically adjusting the knowledge graph by using the learning state feature to obtain a dynamic knowledge graph, comprising the following steps,

[0079] According to the structure of the knowledge points of the electrical automation control course, a knowledge graph with knowledge point nodes, node weights and the association relationship between nodes is constructed. Specifically, each knowledge point in the course is defined as a node of the knowledge graph; the edges between the nodes are established according to the logical relationship (such as prerequisite relationship, correlation degree) between the knowledge points; the initial weight of each knowledge point node is set to reflect its importance or basis.

[0080] The mastery of the students on each knowledge point is analyzed through the learning state features, and the weights of the knowledge point nodes in the knowledge graph are initialized.

[0081] Specifically, the weights of the knowledge point nodes in the knowledge graph are adjusted in real time according to the learning state features of the students, which is represented as,

[0082]

[0083] Wherein, m i (t) is the weight of the i-th knowledge point node at time t, m i (t-1) is the weight of the i-th knowledge point node at time t-1, e is the base of natural logarithm, η is the decay rate parameter, t is the time, λ is the proportion factor of weight update, ReLU is the linear rectifier function, is the learning state feature vector at time t, and is the weight vector related to the i-th knowledge point.

[0084] According to the mastery of the knowledge points, the association strength between the knowledge point nodes is dynamically adjusted, which is represented as,

[0085]

[0086] Wherein, is the association strength between the knowledge point node i and the knowledge point node j at time t, is the initial association strength, tanh is the hyperbolic tangent function, Δm i is the weight change of the i-th knowledge point node, Δm j is the weight change of the j-th knowledge point node, d ij is the distance between the knowledge point node i and the knowledge point node j at time t, and β is the adjustment parameter.

[0087] The adjusted weights and association relationships of the knowledge point nodes are integrated to generate a dynamic knowledge graph that reflects the learning state of the students in real time. Specifically, if the students understand the related knowledge points after learning a certain knowledge point, the association strength between them is enhanced; if the students do not master a certain knowledge point well, the association strength between it and the subsequent knowledge points is weakened.

[0088] It should be noted that by dynamically adjusting the node weight and edge weight in the knowledge graph, the knowledge graph can reflect the learning state and knowledge mastery of the student in real time. The construction and adjustment of the dynamic knowledge graph make the personalized learning path of the student more accurate, and at the same time help the teacher to understand the weak points of the student in real time.

[0089] S4, generating a personalized learning path based on the dynamic knowledge graph, and recommending learning resources, recording learning resource utilization, including the following steps,

[0090] Referring to the node association relationship in the knowledge graph, the adjacency matrix traversal algorithm is used to locate the knowledge points associated with the low weight nodes as review content to generate the learning path planning condition. Specifically, referring to the association relationship between the knowledge points in the dynamic knowledge graph, the adjacency matrix traversal algorithm is used to locate the knowledge points associated with the low weight nodes; the low weight knowledge points and the knowledge points associated therewith are taken as the key content for review and learning, and the learning path planning condition is generated.

[0091] The Dijkstra shortest path algorithm (Dijkstra algorithm) is used to generate a personalized learning path according to the learning path planning condition. Specifically, the Dijkstra shortest path algorithm is used to generate the best personalized learning path for each student according to the path planning condition, so as to ensure that the path is the shortest and covers all the knowledge points to be learned.

[0092] The learning needs of each knowledge point in the personalized learning path are analyzed, and the content-based collaborative filtering recommendation algorithm is used to match the learning resources from the teaching resource library according to the learning needs. Specifically, the learning needs (such as experimental operation guide, theoretical explanation video) of each knowledge point in the personalized learning path are analyzed; the content-based collaborative filtering recommendation algorithm is used to match the learning resources from the teaching resource library, and the resources are sorted according to the relevance and quality.

[0093] The access frequency, learning duration and completion degree of the student to the recommended resources are recorded and analyzed to obtain the resource utilization. Specifically, the use of the recommended resources by the student is monitored, including the access frequency, learning duration and completion degree; according to the monitoring data, the utilization rate of the learning resources is evaluated to provide a basis for subsequent learning state evaluation.

[0094] It should be noted that the dynamic knowledge graph and the recommendation algorithm can be combined to intelligently generate a personalized learning path and accurately recommend learning resources. By optimizing the learning path and resource recommendation, the present application can significantly improve the learning efficiency and knowledge point mastery of the student.

[0095] S5, generating a comprehensive learning state of the student based on the learning state features and the learning resource utilization, including the following steps,

[0096] For each knowledge point node in the dynamic knowledge graph, the weight of the current knowledge point node is extracted. Specifically, the current weight of each knowledge point is extracted from the dynamic knowledge graph, the weight reflects the mastery of the knowledge point by the student, the knowledge point with higher weight indicates that the student has better mastery, and the knowledge point with lower weight indicates that the mastery is poor; the weight is derived from the real-time adjustment of the dynamic knowledge graph, which has considered the learning behavior and state of the student.

[0097] Using the weight of the current knowledge point node, the behavior characteristics in the student learning state are analyzed, and the mastery degree index of the student on the knowledge point in the current learning period is obtained. Specifically, the behavior characteristics include the operation accuracy, experiment completion time and resource usage frequency of the student in the experiment and simulation operation; for each knowledge point node, it is judged whether the student has completed the related experiment task or learning task, and the completion will directly affect the mastery degree index of the knowledge point.

[0098] According to the learning resource utilization rate, the knowledge point mastery index is corrected, and combined with the learning state characteristics, the comprehensive learning state of the student is obtained. Specifically, if the student frequently accesses the resources related to a knowledge point and completes the learning, the mastery index of the knowledge point is improved; if the student uses less resources for a knowledge point or does not complete the related task, the mastery index of the knowledge point is reduced. The comprehensive learning state reflects the overall mastery of the student on most of the knowledge points of the course; the comprehensive learning state can be qualitatively expressed as "excellent", "good", "need to strengthen" and the like, and can also be quantitatively expressed as a comprehensive score or a percentage.

[0099] It should be noted that by extracting the weight of the knowledge point node, analyzing the learning behavior characteristics, and correcting the mastery degree by combining the resource utilization rate, the learning state of the student can be comprehensively reflected. By generating the comprehensive learning state, the overall learning effect of the student can be accurately evaluated, and a reliable basis is provided for subsequent teaching feedback and strategy optimization.

[0100] S6, generating teaching feedback according to the comprehensive learning state, including the following steps,

[0101] According to the comprehensive characteristics of the knowledge point mastery degree and the resource utilization rate, the learning effect of the student is calculated by using a weight distribution algorithm.

[0102] Specifically, the correlation between the knowledge point nodes in the dynamic knowledge graph is used to analyze the contribution degree of different knowledge points to the overall learning effect.

[0103] Specifically, the priority of the key knowledge points is determined through the structure of the knowledge graph, and the mastery of the knowledge points by the student is summarized according to the weight data of the extracted knowledge point nodes and the priority of the knowledge points, to generate the overall learning effect of the student. Specifically, the weight data of the knowledge points is extracted, and the contribution of different knowledge points to the overall learning effect is analyzed; the priority of the key knowledge points is determined, for example, the contribution of the basic knowledge points is higher than that of the advanced knowledge points; the mastery of the knowledge points by the student is summarized, and the evaluation result (such as “excellent”, “good”, “need to strengthen”) of the overall learning effect of the student is generated.

[0104] According to the distribution of the knowledge point mastery degree in the comprehensive learning state, the knowledge points not mastered and poorly mastered by the student are identified. Specifically, the knowledge points with low weight or weak correlation strength of the knowledge point nodes are found out; whether these knowledge points belong to the key nodes in the learning path is analyzed, and the learning problem of the key knowledge points is solved preferentially.

[0105] Based on the identified knowledge points not mastered and poorly mastered by the student, learning resources are pushed through a dynamic recommendation algorithm. Specifically, for the knowledge points not mastered and poorly mastered, relevant experimental guidance, video explanation and exercise sets are recommended; according to the adaptation degree and historical utilization rate of the resources, high-quality resources are preferentially recommended.

[0106] After the pushing is completed, the learning data is re-collected and fed back, and the dynamic knowledge graph and the personalized learning path are updated. Specifically, the weight of the knowledge point nodes in the knowledge graph is updated to reflect the latest mastery of the knowledge points by the student; the correlation strength between the knowledge points is adjusted to optimize the logical relationship of the knowledge points; according to the updated dynamic knowledge graph, the personalized learning path of the student is re-planned to ensure that the learning path can cover all the not mastered knowledge points and meet the review needs.

[0107] It should be noted that by generating targeted teaching feedback and optimizing teaching strategies, the weak links in the student's learning can be effectively solved; by dynamically recommending learning resources and updating the knowledge graph, a closed-loop learning optimization system is formed. By generating teaching feedback and optimizing teaching strategies, the learning effect and teaching efficiency of the student can be significantly improved, and strong support for personalized teaching is provided.

[0108] The embodiment also provides a teaching information processing system based on electrical automation control, comprising: an information collection module, a feature extraction module, a knowledge graph module, a learning resource module, a learning state module and a feedback module.

[0109] The information collection module is configured to collect multi-modal teaching information and perform preprocessing; the feature extraction module is configured to extract and fuse features by using the preprocessed multi-modal teaching information to generate learning state features; the knowledge graph module is configured to construct and dynamically adjust a knowledge graph by using the learning state features to obtain a dynamic knowledge graph; the learning resource module is configured to generate a personalized learning path based on the dynamic knowledge graph, recommend learning resources, and record learning resource utilization; the learning state module is configured to generate a comprehensive learning state of a student based on the learning state features and the learning resource utilization; and the feedback optimization module is configured to generate teaching feedback according to the comprehensive learning state.

[0110] The embodiment also provides a computer device suitable for the case of the teaching information processing method based on electrical automation control, including a memory and a processor; the memory is configured to store computer executable instructions, and the processor is configured to execute the computer executable instructions to implement the teaching information processing method based on electrical automation control proposed in the above embodiment.

[0111] The computer device can be a terminal, and the computer device includes a processor, a memory, a communication interface, a display screen and an input device connected through a system bus. The processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operating system and the computer program in the non-volatile storage medium to run. The communication interface of the computer device is configured to perform wired or wireless communication with external terminals. The wireless communication can be achieved through WIFI, operator network, NFC (near field communication) or other technologies. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen. The input device of the computer device can be a touch layer overlaid on the display screen, or a key, trackball or touchpad arranged on the shell of the computer device. In addition, the input device can also be an external keyboard, touchpad or mouse, etc.

[0112] The embodiment also provides a storage medium on which a computer program is stored, the program being executed by a processor to implement the teaching information processing method based on electrical automation control proposed in the above embodiment; the storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as a static random access memory (SRAM), an electrically erasable programmable read-only memory (EEPROM), an erasable programmable read-only memory (EPROM), a programmable read-only memory (PROM), a read-only memory (ROM), a magnetic memory, a flash memory, a magnetic disk, or an optical disk.

[0113] To sum up, the present application dynamically evaluates the knowledge point mastery of students by constructing a dynamic knowledge graph, combining the learning state characteristics of students and the learning resource utilization rate, generates a personalized learning path and pushes adaptive learning resources. The present application can update the weights and the association relationship of each knowledge point node in the knowledge graph in real time, reflect the learning progress and weak links of students, thereby dynamically adjusting the teaching strategy to improve the pertinence and individualization of teaching, through the dynamic knowledge graph, a student's exclusive learning path is constructed, and the learning efficiency is optimized. A real-time feedback mechanism is provided to help students quickly find weak points in learning, and the mastery of knowledge points is improved through the accurate recommendation of adaptive resources. The closed-loop management of learning data is realized, and through the collection, analysis and feedback of student learning data, the teaching resources and path planning are continuously optimized.

[0114] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present application but not limit the present application. Although the present application has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present application can be modified or replaced by equivalents without departing from the spirit and scope of the present application, and all modifications or replacements should be covered in the scope of the claims of the present application.

Claims

1. A teaching information processing method based on electrical automation control, characterized in that: include, Collect multimodal teaching information and perform preprocessing; Utilize the pre-processed multimodal teaching information to extract and fuse features to generate learning state features; Utilize learning state features to construct and dynamically adjust the knowledge graph to obtain a dynamic knowledge graph; Generate personalized learning paths based on dynamic knowledge graphs, recommend learning resources, and record learning resource utilization; Generate students' comprehensive learning status based on learning status characteristics and learning resource utilization; Generate teaching feedback based on comprehensive learning status; Use learning state features to construct and dynamically adjust the knowledge graph to obtain a dynamic knowledge graph. The specific steps are as follows: According to the structure of knowledge points in the electrical automation control course, a knowledge graph with knowledge point nodes, node weights, and relationships between nodes is constructed; Analyze students' mastery of each knowledge point through learning status characteristics and initialize the weights of knowledge point nodes in the knowledge graph; According to the learning status characteristics of students, the weight of each knowledge point node in the knowledge graph is adjusted in real time, which is expressed as: ； in, For the Knowledge point nodes at time The weight of For the Knowledge point nodes at time The weight of is the base of natural logarithms, is the decay rate parameter, For time, is the scaling factor for weight update, is a linear rectification function, For time The learning state feature vector of For the The weight vector associated with each knowledge point; According to the mastery of knowledge points, the association strength between each knowledge point node is dynamically adjusted, which is expressed as: ； in, For in time Knowledge point node and knowledge points The strength of the association between is the initial correlation strength, is the hyperbolic tangent function, For the The weight change of each knowledge point node, For the The weight change of each knowledge point node, For time Knowledge point node and knowledge point nodes The distance between is the adjustment parameter; The adjusted weights and associations of knowledge point nodes are integrated to generate a dynamic knowledge graph that reflects students' learning status in real time.

2. The teaching information processing method based on electrical automation control according to claim 1, characterized in that: The multimodal teaching information includes students' operation data, measurement data and analysis results, experimental report content, homework completion status, answer records, study notes, study time and frequency of access to learning resources in experiments and simulation operations of the electrical automation control course; The preprocessing includes data cleaning, denoising, formatting and spatiotemporal alignment.

3. The teaching information processing method based on electrical automation control according to claim 2, characterized in that: Using the pre-processed multimodal teaching information, extract and fuse features to generate learning state features. The specific steps are as follows: Use convolutional neural networks to extract features from operational data, measurement data, and analysis results in experimental and simulation operations to obtain behavioral features; Analyze lab reports, homework, and study notes using the BERT natural language processing method to extract text features; The behavioral features and text features are fused through the multimodal attention mechanism. Before fusion, the attention weight needs to be calculated. The calculation method is expressed as: , ; in, is the attention weight of behavioral features to text features, is a function that converts an input vector into a probability distribution, For behavioral characteristics, is the text feature, is the pre-linear transformation matrix, is the post linear transformation matrix, is the attention weight of text features to behavioral features, is the normalization factor; The behavioral features and text features are integrated through the multimodal attention mechanism to obtain the student's learning state features, which can be expressed as: ; in, The characteristics of students’ learning status, is a linear rectification function, is the attention weight of behavioral features to text features, is the text feature, is the attention weight of text features to behavioral features, For behavioral characteristics.

4. The teaching information processing method based on electrical automation control according to claim 3, characterized in that: Generate personalized learning paths based on dynamic knowledge graphs, recommend learning resources, and record learning resource utilization. The specific steps are as follows: Referencing the node association relationships in the knowledge graph, an adjacency matrix traversal algorithm is used to locate knowledge points associated with low-weight nodes as review content, generating conditions for learning path planning; Generate personalized learning paths using Dijkstra's shortest path algorithm and based on learning path planning conditions; Analyze the learning needs of each knowledge point in the personalized learning path, and use the content-based collaborative filtering recommendation algorithm to match appropriate learning resources from the teaching resource library based on the learning needs; Record and analyze students' access frequency, learning time and completion rate of recommended resources to obtain resource utilization rate.

5. The teaching information processing method based on electrical automation control according to claim 4, characterized in that: Based on the learning status characteristics and learning resource utilization rate, the comprehensive learning status of students is generated. The specific steps are as follows: For each knowledge point node in the dynamic knowledge graph, extract the weight of the current knowledge point node; Utilize the weight of the current knowledge point node to analyze the behavioral characteristics of the student's learning status and obtain the student's mastery of the knowledge point in the current learning cycle; According to the utilization rate of learning resources, the knowledge point mastery index is revised and combined with the learning status characteristics to obtain the student's comprehensive learning status.

6. The teaching information processing method based on electrical automation control according to claim 5, characterized in that: Generate teaching feedback based on the comprehensive learning status. The specific steps are as follows: Based on the comprehensive characteristics of knowledge point mastery and resource utilization, a weight distribution algorithm is used to calculate the overall learning effect of students; Utilize the correlation between knowledge point nodes in the dynamic knowledge graph to analyze the contribution of different knowledge points to the overall learning effect; The structure of the knowledge graph determines the priority of key knowledge points, and based on the weight data of the extracted knowledge point nodes and the priority of the knowledge points, summarizes the students' mastery of the knowledge points to generate the students' overall learning effect; According to the distribution of knowledge points mastery in the comprehensive learning status, identify the knowledge points that students have not mastered or have mastered poorly; Based on the knowledge points that students have not mastered or have mastered poorly, learning resources are pushed through dynamic recommendation algorithms; After the push is completed, the learning data will be re-collected and fed back to update the dynamic knowledge graph and personalized learning path.

7. A teaching information processing system based on electrical automation control, based on the teaching information processing method based on electrical automation control according to any one of claims 1 to 6, characterized in that: Including information acquisition module, feature extraction module, knowledge graph module, learning resource module, learning status module and feedback module; The information acquisition module is used to collect multimodal teaching information and perform preprocessing; The feature extraction module is used to extract and fuse features using the pre-processed multimodal teaching information to generate learning state features; The knowledge graph module is used to construct and dynamically adjust the knowledge graph using learning state features to obtain a dynamic knowledge graph; The learning resource module is used to generate personalized learning paths based on the dynamic knowledge graph, recommend learning resources, and record the utilization rate of learning resources; The learning status module is used to generate a comprehensive learning status of the student based on the learning status characteristics and learning resource utilization rate; The feedback module is used to generate teaching feedback based on the comprehensive learning status.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the teaching information processing method based on electrical automation control according to any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the teaching information processing method based on electrical automation control according to any one of claims 1 to 6 are implemented.

Images