English course management system applied to English teaching

Abstract

The invention relates to an English course management system applied to English teaching, and the system comprises the steps: collecting multi-source English teaching data, carrying out the cleaning and feature extraction, and constructing a knowledge graph according to the extracted feature parameters; calculating curriculum configuration parameter factors by utilizing a feature filtering and weighting formula based on the knowledge graph; inputting the course configuration parameter factors into a deep reinforcement learning model for strategy training; outputting an optimal course content and difficulty configuration combination in the current teaching environment, and dynamically updating a teaching strategy; according to the scheme, a knowledge graph, multi-feature weighting and deep reinforcement learning combined mode is introduced, the complex feature relation in the English teaching process is accurately captured, the scheme can be evolved continuously through the self-adaptive optimization capability of reinforcement learning, and therefore the accuracy and flexibility of teaching resource allocation and course difficulty regulation and control are remarkably improved.

Description

Technical Field

[0001] The invention belongs to the technical field of course resource management, and in particular relates to an English course management system applied to English teaching. Background Art

[0002] In today's globalized context, English is the universal language for international communication, and its teaching and course management are highly valued by educational institutions and learners. With the rapid development of Internet technology, multimedia resources and artificial intelligence algorithms, English teaching models based on networks and intelligent platforms have become mature and diversified. Traditional English teaching management systems mostly rely on pre-set syllabi and fixed teaching materials, and are mainly based on linear and one-way teaching methods. However, with the popularization of online education models, multi-source heterogeneous teaching resources (such as audio, video, e-books, online assessment tools, and corpora in multicultural backgrounds) are becoming more and more abundant, and students' personalized needs in the process of English learning are becoming more and more complex. This makes it difficult for teaching plans based solely on rules or experience to adapt to diverse teaching situations. In the prior art, some studies have attempted to use data mining and big data analysis methods to assist in the formulation of teaching strategies, but they are mostly based on static feature extraction and simple weighted analysis, and it is difficult to make full use of semantic associations and dynamic feedback information across resources. In addition, although some studies have used artificial intelligence to intelligently allocate teaching resources, there are still deficiencies in feature selection, correlation measurement, and dynamic strategy optimization, and they have failed to effectively build an adaptive and evolving course management model.

[0003] Under the background of existing related technologies, the core problems are: first, in the management of multi-source English teaching resources, there is a lack of deep mining of the semantics and correlation between each feature, resulting in a deviation between the selected teaching resources and the actual needs of students; second, the update of teaching strategies is mostly phased and passive, and there is a lack of intelligent learning strategy optimization methods that can be continuously adjusted according to real-time feedback. These deficiencies make it difficult for the system to continuously provide learners with optimal and personalized learning paths and resource configurations. Based on this, our invention proposes to use knowledge graphs to construct feature parameter associations, thereby establishing an organic connection between different types of English teaching resources, and dynamically adjust strategies with the help of deep reinforcement learning models. Through the construction of feature filtering and weighted formulas, the quantitative integration of multi-dimensional feature information is realized, and the decision is continuously updated in the cyclic training of deep reinforcement learning, so that the system can adaptively generate the optimal course configuration plan in different teaching situations. The existing English course management technology has the problem of insufficient resource allocation and strategy optimization, while the present invention solves the defects of the existing technology in dynamic association mining and strategy adaptive optimization through a method combining knowledge graphs with deep reinforcement learning. The present invention belongs to the field of intelligent English teaching and course resource management technology. Summary of the invention

[0004] The purpose of this section is to summarize some aspects of embodiments of the present invention and briefly introduce some preferred embodiments. Some simplifications or omissions may be made in this section and the specification abstract and the invention title of this application to avoid blurring the purpose of this section, the specification abstract and the invention title, and such simplifications or omissions cannot be used to limit the scope of the present invention.

[0005] In view of the following technical problems existing in the prior art: the defects of the prior art in dynamic association mining and strategy adaptive optimization are solved.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions: an English course management system applied to English teaching, characterized in that: it includes:

[0007] Collect multi-source English teaching data, clean it and extract features, and build a knowledge graph based on the extracted feature parameters;

[0008] Calculate the course configuration parameter factors using feature filtering and weighted formula based on knowledge graph;

[0009] Inputting the course configuration parameter factors into a deep reinforcement learning model for strategy training;

[0010] Output the optimal course content and difficulty configuration combination under the current teaching environment and dynamically update the teaching strategy.

[0011] As a preferred technical solution for an English course management system applied to English teaching, the cleaning and feature extraction steps include normalizing the original data and removing outliers and redundant features to obtain a standardized feature vector set.

[0012] As an optimal technical solution for an English course management system applied to English teaching, the following formulas are used to simultaneously demonstrate and illustrate the feature filtering and weighting steps:

[0013]

[0014] Among them, F is the course configuration parameter factor;

[0015] V i is the eigenvalue of the i-th feature;

[0016] W i is the weight parameter of the corresponding feature, which is determined by the relevance in the knowledge graph and the priority of the syllabus;

[0017] R i is the feature association coefficient, which is dynamically calculated based on the semantic relationship between nodes in the knowledge graph;

[0018] The course configuration parameter factor is used to quantify the match between the current student learning status and teaching resources.

[0019] As an optimal technical solution for an English course management system applied to English teaching, the course configuration parameter factors are used as the state input of the deep reinforcement learning model, and the strategy optimization is achieved through the following steps in each training iteration:

[0020] Input the neural network model according to the current state and output the action strategy;

[0021] Calculate reward value R based on student learning feedback w ,in:

[0022] R w =α·ΔS-β·C

[0023] Among them, ΔS is the improvement value of students' test scores, C is the resource adjustment cost, and α and β are balance coefficients;

[0024] Using the reward value R w Update the policy network parameters so that the policy is more inclined to choose actions that can improve students' learning outcomes in subsequent iterations.

[0025] As an optimal technical solution for an English course management system applied to English teaching, the correlation between nodes in the knowledge graph is continuously updated during the training process, so that the feature correlation coefficient R i It can adapt to changes, thus achieving personalized customization for different students in the long-term teaching process.

[0026] As an optimal technical solution for an English course management system applied to English teaching, in the step of outputting the optimal course configuration combination, the finalized course difficulty level, resource type allocation and question type structure are presented to teachers and students, and dynamically adjusted regularly based on students' learning performance.

[0027] As an optimal technical solution for an English course management system applied to English teaching, by periodically recalculating the feature parameter weight W i Correlation coefficient R i , adjust the course parameter factor F in time when the student's learning progress changes, ensuring that the course configuration is applicable at different teaching stages

[0028] Beneficial effects of the invention: This solution introduces a combination of knowledge graphs, multi-feature weighting and deep reinforcement learning, which not only accurately captures the complex feature relationships in the English teaching process, but also enables the solution to evolve continuously through the adaptive optimization capability of reinforcement learning, thereby significantly improving the accuracy and flexibility of teaching resource allocation and course difficulty control. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing 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 creative labor. Among them:

[0030] Figure 1 It is a schematic diagram of the flow structure of the system method of the present invention; DETAILED DESCRIPTION

[0031] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings.

[0032] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and 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.

[0033] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0034] Secondly, the present invention is described in detail with reference to the schematic diagram. When describing the embodiments of the present invention in detail, for the sake of convenience, the cross-sectional diagrams showing the device structure will not be partially enlarged according to the general scale, and the schematic diagrams are only examples, which should not limit the scope of protection of the present invention. In addition, in actual production, the three-dimensional dimensions of length, width and depth should be included.

[0035] Example 1

[0036] Reference Figure 1 This embodiment provides an English course management system for English teaching, including:

[0037] Collect multi-source English teaching data, clean it and extract features, and build a knowledge graph based on the extracted feature parameters;

[0038] Calculate the course configuration parameter factors using feature filtering and weighted formula based on knowledge graph;

[0039] Inputting the course configuration parameter factors into a deep reinforcement learning model for strategy training;

[0040] Output the optimal course content and difficulty configuration combination under the current teaching environment and dynamically update the teaching strategy.

[0041] The cleaning and feature extraction steps include normalizing the original data (including student test scores, study time, error rate, and learning resource click records) and removing outliers and redundant features to obtain a standardized feature vector set.

[0042] In the feature filtering and weighting steps, the following formulas are used to demonstrate and explain them simultaneously:

[0043]

[0044] Among them, F is the course configuration parameter factor;

[0045] V i is the characteristic value of the i-th characteristic, such as the student's mastery score of a certain type of resource;

[0046] W i is the weight parameter of the corresponding feature, which is determined by the relevance in the knowledge graph and the priority of the syllabus;

[0047] R i is the feature association coefficient, which is dynamically calculated based on the semantic relationship between nodes in the knowledge graph (such as textbook chapter association and question type similarity);

[0048] The F value calculated by this formula is used to quantify the match between the current student learning status and teaching resources.

[0049] The course configuration parameter factors are used as the state input of the deep reinforcement learning model (such as the Actor-Critic structure based on the policy gradient method), and the policy optimization is achieved through the following steps in each training iteration:

[0050] Input the neural network model according to the current state and output action strategies (such as increasing the difficulty of the course, increasing or decreasing the listening training materials, and adjusting the reading order of the text);

[0051] Calculate reward value R based on student learning feedback w ,in:

[0052] R w =α·ΔS-β·C

[0053] Among them, ΔS is the improvement value of students' test scores, C is the resource adjustment cost, and α and β are balance coefficients;

[0054] Using the reward value R w Update the policy network parameters so that the policy is more inclined to choose actions that can improve students' learning outcomes in subsequent iterations.

[0055] During the training process, the correlation between nodes in the knowledge graph is continuously updated so that the feature correlation coefficient R i It can adapt to changes, thus achieving personalized customization for different students in the long-term teaching process.

[0056] In the step of outputting the optimal course configuration combination, the finalized course difficulty level, resource type allocation (such as the ratio of the four types of resources: listening, speaking, reading, and writing) and question type structure are presented to teachers and students, and dynamically adjusted regularly based on students' learning performance.

[0057] By periodically recalculating the feature parameter weights W i Correlation coefficient R i , the course parameter factor F is adjusted in time when the students' learning progress changes, ensuring that the course configuration is applicable in different teaching stages.

[0058] Example description:

[0059] Step 1: Data collection and cleaning (proceed to step 2 after completion)

[0060] Collect raw data from multiple sources (such as online learning platform access logs, assessment system score records, and oral assessment data annotated by teachers). Format these data uniformly, delete or correct abnormal data (such as abnormal login time, extreme assessment results), and use normalization algorithms to map data of different dimensions to the [0,1] interval to form a standardized feature vector set.

[0061] Step 2: Feature filtering and weighting (after completion, proceed to step 3)

[0062] Based on knowledge graph construction: establish association models for different types of teaching resources (such as word lists, listening materials, reading texts), student learning characteristics (such as vocabulary, oral fluency, reading speed) and teaching objectives (such as IELTS, TOEFL, ordinary high school English courses). Determine the weight of each feature by analyzing the edge weights and relevance between nodes and setting the priority of the syllabus. i and the correlation coefficient R i , and then the relevant eigenvalue V i Substitute into the formula F = ∑ i (V i ·W i ·R i ) Calculate the course configuration parameter factor F. The higher the F value, the more conducive the current combination of specific features is to improving students’ English proficiency.

[0063] Step 3: Deep reinforcement learning strategy training (after completion, proceed to step 4)

[0064] The FFF value is used as the state input into the DRL model (such as based on the Actor-Critic framework). The Actor network outputs the strategy based on the state, and the Critic network evaluates the value generated by the strategy. In each round of training, the reward R is calculated based on the student's learning performance (such as the improvement value ΔS of the new round of evaluation scores). w =α·ΔS-β·C, and use the reward to update the neural network parameters. Through multiple rounds of iterations, the policy network continues to converge and eventually obtains the optimal teaching strategy.

[0065] Step 4: Output and Dynamic Update

[0066] After the training is completed, the course configuration (such as the number of questions, difficulty, and resource type allocation) corresponding to the optimal strategy is output to the system to provide students with personalized course content. At the same time, in subsequent teaching, as the learning characteristics of students change, the above steps 1 to 3 are repeated regularly to recalculate the knowledge graph relevance and feature parameters to achieve continuous optimization of the teaching plan.

[0067] By collecting multi-source English teaching data and performing cleaning and feature extraction, a knowledge graph is constructed based on the extracted feature parameters, which enables the extraction of feature elements with clear correlations from massive, heterogeneous and messy teaching data, and the structured presentation of various teaching resources, learning behavior data and related semantic information in the form of a knowledge graph. The effect of this is that the system no longer makes decisions based solely on linear or single-dimensional features, but has the ability to express resource relationships in a multi-dimensional and hierarchical manner. Through this process, the beneficial effect of providing a high-quality, multi-angle feature input foundation for subsequent strategy optimization links is ultimately achieved, effectively ensuring the comprehensiveness and accuracy of decision-making in subsequent steps.

[0068] By using feature filtering based on the knowledge graph and weighted formula to calculate the course configuration parameter factors, the process of eliminating redundant features, amplifying key features, and comprehensively considering related features is achieved. In this step, the system does not rely on experience or simple statistical rules, but quantifies key parameters by introducing the correlation between nodes in the knowledge graph and the feature weight distribution method. The role is to condense multi-source feature information into quantifiable and comparable parameter factors to ensure that the subsequent deep reinforcement learning process does not "lose direction" due to excessive redundant information. Ultimately, the beneficial effects of this process are reflected in the significant improvement of the input quality and computational efficiency of the decision-making model, laying a solid foundation for precise focus and rapid convergence in the strategy training process.

[0069] By inputting the course configuration parameter factors into the deep reinforcement learning model for strategy training, the autonomous iteration and dynamic optimization of the teaching strategy are achieved. In this process, the deep reinforcement learning model no longer stops at static strategy calculation, but continuously corrects and evolves the strategy parameters through continuous decision feedback loops. Its role is to enable the system to maintain accurate adjustment and improvement of strategies in real teaching scenarios when faced with the variability of students' learning behavior patterns and the dynamic changes of diversified teaching resources. The ultimate beneficial effect is that the model shows extremely high adaptability and resilience to complex teaching environments, ensuring that the system continues to provide the most targeted and optimally configured teaching plans for different learners in long-term use.

[0070] By outputting the optimal course content and difficulty configuration combination under the current teaching environment and dynamically updating the teaching strategy, the immediate implementation of personalized and adaptive teaching plans is achieved. This step is no longer simply to provide a set of fixed plans, but to use the strategies optimized by deep reinforcement learning in the actual teaching process, and continuously update them as the environment, student feedback and performance change. Its role is that the system can re-analyze the evaluation at the end of each learning cycle, so as to quickly iterate a better course combination in the next cycle. Ultimately, the beneficial effect of this process is to enable the teaching system to have a lasting evolutionary ability, continuously improve learning effects and teaching efficiency, meet the dynamic needs of different student groups in the long-term learning process, and realize a highly intelligent and adaptive English course management system.

[0071] In summary, through the progressive and linkage of the above steps, the present invention has accumulated and embodied significant beneficial effects in multi-source data processing, feature weight quantification, deep strategy optimization, and dynamic teaching program output. These steps promote each other, from high-quality structured representation of input data, to accurate quantification of key parameters, to continuous evolution of deep learning strategy models, and finally to intelligent dynamic management of the actual teaching process, achieving the overall beneficial effects of improving the utilization rate of English teaching resources, optimizing learning strategies, and improving learners' learning efficiency.

[0072] This solution introduces a combination of knowledge graphs, multi-feature weighting and deep reinforcement learning. It not only accurately captures the complex feature relationships in the English teaching process, but also enables the solution to evolve continuously through the adaptive optimization capabilities of reinforcement learning, thereby significantly improving the accuracy and flexibility of teaching resource allocation and course difficulty control.

[0073] It will be appreciated that in the development of any actual implementation, as in any engineering or design project, numerous implementation-specific decisions may be made. Such a development effort may be complex and time-consuming, but will be a routine task of design, fabrication, and production for those of ordinary skill having the benefit of this disclosure without undue experimentation.

[0074] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention 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 invention can be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should be included in the scope of the claims of the present invention.

Claims

1. An English course management system for English teaching, characterized by: include, Collect multi-source English teaching data, clean it and extract features, and build a knowledge graph based on the extracted feature parameters; Calculate the course configuration parameter factors using feature filtering and weighted formula based on knowledge graph; Inputting the course configuration parameter factors into a deep reinforcement learning model for strategy training; Output the optimal course content and difficulty configuration combination under the current teaching environment and dynamically update the teaching strategy.

2. The English course management system for English teaching according to claim 1, characterized in that: The cleaning and feature extraction steps include normalizing the original data and removing outliers and redundant features to obtain a standardized feature vector set.

3. The English course management system for English teaching according to claim 2, characterized in that: In the feature filtering and weighting steps, the following formulas are used to demonstrate and explain them simultaneously: Among them, F is the course configuration parameter factor; V i is the eigenvalue of the i-th feature; W i is the weight parameter of the corresponding feature, which is determined by the relevance in the knowledge graph and the priority of the syllabus; R i is the feature association coefficient, which is dynamically calculated based on the semantic relationship between nodes in the knowledge graph; The course configuration parameter factor is used to quantify the match between the current student learning status and teaching resources.

4. The English course management system for English teaching according to claim 3 is characterized by: The curriculum configuration parameter factors are used as the state input of the deep reinforcement learning model, and the policy optimization is achieved through the following steps in each training iteration: Input the neural network model according to the current state and output the action strategy; Calculate reward value R based on student learning feedback w ,in: R w =α·ΔS-β·C Among them, ΔS is the improvement value of students' test scores, C is the resource adjustment cost, and α and β are balance coefficients; Using the reward value R w Update the policy network parameters so that the policy is more inclined to choose actions that can improve students' learning outcomes in subsequent iterations.

5. The English course management system for English teaching according to claim 4 is characterized in that: During the training process, the correlation between nodes in the knowledge graph is continuously updated so that the feature correlation coefficient R i It can adapt to changes, thus achieving personalized customization for different students in the long-term teaching process.

6. The English course management system for English teaching according to claim 5, characterized in that: In the step of outputting the optimal course configuration combination, the finalized course difficulty level, resource type allocation and question type structure are presented to teachers and students, and dynamically adjusted regularly based on students' learning performance.

7. The English course management system for English teaching according to claim 6, characterized in that: By periodically recalculating the feature parameter weights W i Correlation coefficient R i , the course parameter factor F is adjusted in time when the students' learning progress changes, ensuring that the course configuration is applicable in different teaching stages.

Images