Intelligent English teaching method and system and storage medium

Abstract

The invention belongs to the technical field of English teaching methods, and particularly relates to an intelligent English teaching method and system and a storage medium, and the method comprises the steps: constructing a multi-dimensional linguistic feature analysis model, and extracting lexical features, syntactic relationship features and semantic deviation features in a text input by a student through a natural language processing technology; dynamic student portraits are established based on the cognitive psychology theory, cognitive level labels are updated according to real-time learning data, and the data comprise grammar error clustering distribution, spoken language fluency indexes and vocabulary association response time; and generating a personalized teaching path, matching teaching materials from the hierarchical resource library according to the cognitive level label, and dynamically adjusting the complexity and presentation form of a teaching strategy. Lexical, syntactic and semantic deviation features are extracted through a natural language processing technology, student error types can be accurately positioned, the problem that traditional error correction only stays on surface modification is avoided, and cognitive tags are updated based on real-time learning data.

Description

Technical Field

[0001] This invention belongs to the field of English teaching methods technology, specifically relating to an intelligent English teaching method, system, and storage medium. Background Technology

[0002] The following technical pain points are currently prevalent in the field of English teaching:

[0003] The teaching is highly homogenized: traditional teaching models are unable to provide precise guidance for individual student differences. Although existing intelligent teaching systems can correct basic grammar errors, they lack in-depth analysis of the root causes of errors, cannot form systematic improvement strategies, and have insufficient dynamic adaptability. Most online learning platforms adopt fixed difficulty levels and fail to dynamically adjust teaching content based on students' real-time learning performance.

[0004] The lack of cross-cultural teaching: Current technologies focus on language form training, neglecting the impact of cultural differences on language acquisition. For example, native Chinese speakers often make subject-verb disagreement errors in English writing due to a "topic-first" thinking pattern, while traditional systems only mark the errors without providing cultural comparison analysis.

[0005] Delayed evaluation and feedback: Teachers rely on manual grading of homework and tests, making it difficult to identify gaps in group knowledge in a timely manner, resulting in teaching adjustments lagging behind changes in students' learning. Summary of the Invention

[0006] To overcome the shortcomings of existing technologies, this invention provides an intelligent English teaching method, system, and storage medium to address the serious homogenization of teaching: traditional teaching models struggle to provide precise guidance tailored to individual student differences; while existing intelligent teaching systems can correct basic grammar errors, they lack in-depth analysis of the root causes of errors, failing to formulate systematic improvement strategies and exhibiting insufficient dynamic adaptability; most online learning platforms use fixed difficulty levels, failing to dynamically adjust teaching content based on students' real-time learning performance; and existing technologies focus on language form training, neglecting the impact of cultural differences on language acquisition. For example, native Chinese speakers often make subject-verb disagreement errors in English writing due to a "topic-first" thinking pattern, while traditional systems only mark the errors without providing cultural comparison analysis; teachers rely on manual grading of assignments and tests, making it difficult to promptly identify group-wide knowledge gaps, resulting in teaching adjustments lagging behind changes in student learning.

[0007] One embodiment of the present invention provides an intelligent English teaching method, comprising the following steps:

[0008] A multi-dimensional linguistic feature analysis model is constructed, and lexical features, syntactic relation features, and semantic deviation features in student input text are extracted using natural language processing technology.

[0009] A dynamic student profile is established based on cognitive psychology theory, and the cognitive level label is updated according to real-time learning data. The data includes the distribution of grammatical errors, oral fluency index and vocabulary association response time.

[0010] It generates personalized learning paths, matches teaching materials from a graded resource library based on cognitive level labels, and dynamically adjusts the complexity and presentation of teaching strategies.

[0011] In one embodiment, the linguistic feature analysis model is implemented through the following steps:

[0012] Transformational generative grammar theory is used to perform deep structural analysis of the target sentence and to annotate part-of-speech dependency relations and semantic roles;

[0013] Knowledge graph technology is used to calculate the semantic similarity between student expressions and standard corpora, and error type diagnostic reports are generated.

[0014] In one embodiment, the construction of the dynamic student profile includes:

[0015] By applying Piaget's theory of cognitive development stages to divide language ability levels, students' error patterns are classified as either symbolic deficiencies in the preoperational stage or logical fallacies in the formal operational stage.

[0016] The forgetting curve algorithm is used to predict the vocabulary memory decay cycle and automatically plan review time nodes.

[0017] In one embodiment, the generation of the personalized learning path includes:

[0018] For students with pronunciation deviations, a linking weakening training module is inserted, and tongue position correction animation is generated through acoustic feature comparison;

[0019] For students whose syntactic error rate is higher than the threshold, we push grammar micro-lesson videos. The video content is based on Gagné's hierarchy of learning theory, progressing step by step from concept learning to rule application.

[0020] In one embodiment, the generation of the grammar micro-lesson video further includes:

[0021] High-frequency sentence patterns are extracted from real-time news corpora, and contextualized example sentences are generated through semantic role labeling.

[0022] Based on cognitive load theory, the information density of a single video is controlled to ensure that the relevance of new knowledge points to students' existing knowledge structure exceeds a preset threshold.

[0023] In one embodiment, a teacher-side collaboration method is also included:

[0024] Automatically generate a class learning heatmap to visually present common weaknesses and the distribution of individual differences;

[0025] Group discussion topics are recommended based on the principles of communicative teaching, and the difficulty of the topics is dynamically balanced according to the cognitive labels of the members.

[0026] One embodiment of the present invention provides an intelligent English teaching system according to the method described in any of the above embodiments, comprising:

[0027] A linguistic feature analysis engine that integrates random forest algorithm and LSTM neural network to process multimodal learning data;

[0028] A cognitive modeling database that stores knowledge units and associated cognitive strategies organized according to Bruner's discovery learning theory.

[0029] In one embodiment, the optimization method of the linguistic feature analysis engine includes:

[0030] Adversarial training techniques are used to improve the generalization ability of dialect accents, and acoustic models are used to transfer learning to adapt to the pronunciation features of the region.

[0031] An attention mechanism is introduced to enhance the accuracy of syntax tree parsing for long and complex sentences.

[0032] In one embodiment, the method for updating the cognitive modeling database includes:

[0033] Regularly collect students' cross-platform learning behavior data and use Bayesian networks to infer the effectiveness of cognitive strategies.

[0034] Reconstruct the knowledge topology based on Ausubel's assimilation theory to ensure that the semantic distance between new concepts and existing anchor points is less than an acceptable range.

[0035] One embodiment of the present invention provides a computer-readable storage medium storing program instructions, which, when executed, implement the intelligent English teaching method described above.

[0036] Compared with the prior art, the embodiments of the present invention have the following beneficial effects:

[0037] 1. A syntactic analysis model based on transformational generative grammar theory can identify complex errors in student essays, such as "logical breaks in nested clauses," and generate attribution reports by associating them with the student's native language background (e.g., misuse of the present perfect tense due to the lack of tense markers in Chinese). Semantic biases can be quantified through knowledge graphs. For example, when a student misuses "originate from" in the description of modern events, the system automatically pushes contextual comparison cases of the phrase with "date backto."

[0038] 2. The dynamic cognitive modeling module, based on Vygotsky's zone of proximal development theory, automatically unlocks higher-order tasks such as "debate-style dialogues" when a student's oral fluency index reaches a threshold, avoiding ineffective repetitive training. For students with pronunciation deviations, the system generates tongue position animations and spectrograms for comparison, visually demonstrating the direction of deviation between their vowel formants and standard pronunciation.

[0039] 3. In writing tutoring, we not only correct grammatical errors, but also explain the causes of errors through visual diagrams of the differences between Chinese and English thinking (such as "English subject-verb priority vs. Chinese topic priority"), thereby strengthening cross-cultural communication skills.

[0040] 4. The teacher-side heatmap can display the cluster distribution of common errors in the class and recommend targeted micro-lesson videos and tiered group activities. The student-side uses the LSTM model to predict the vocabulary forgetting cycle and pushes review materials containing the original learning context (such as news excerpts) at the memory decay threshold to enhance long-term memory. Attached Figure Description

[0041] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the structures shown in these drawings without creative effort.

[0042] Figure 1 This is a schematic diagram of the intelligent English teaching system architecture of the present invention;

[0043] Figure 2 This is a flowchart illustrating the multi-dimensional feature analysis model of the present invention.

[0044] Figure 3 This is a schematic diagram of the personalized teaching path generation interface of the present invention. Detailed Implementation

[0045] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0046] The terms "first," "second," etc., used in the specification, claims, and accompanying drawings of this invention are used to distinguish different objects, not to describe a specific order. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, apparatus, product, or end that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products, or ends.

[0047] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of the invention. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0048] like Figure 1-3 As shown, one embodiment of the present invention provides an intelligent English teaching method, including the following steps:

[0049] A multi-dimensional linguistic feature analysis model is constructed, and lexical features, syntactic relation features, and semantic deviation features in student input text are extracted using natural language processing technology.

[0050] A dynamic student profile is established based on cognitive psychology theory, and the cognitive level label is updated according to real-time learning data. The data includes the distribution of grammatical errors, oral fluency index and vocabulary association response time.

[0051] It generates personalized learning paths, matches teaching materials from a graded resource library based on cognitive level labels, and dynamically adjusts the complexity and presentation of teaching strategies.

[0052] In this embodiment of the invention, lexical, syntactic, and semantic deviation features are extracted using natural language processing technology, which can accurately locate the types of student errors (such as missing articles, tense confusion, etc.), avoiding the problem that traditional error correction only stays at the surface level. Based on real-time learning data (such as spoken fluency and vocabulary association speed), cognitive labels are updated, so that the teaching strategy always matches the student's current ability level, solving the problems of "too difficult and frustrating" or "too simple and inefficient" caused by fixed levels. Combined with a leveled resource library, the complexity of teaching content is dynamically adjusted (such as adding subjunctive mood training for advanced students), so as to achieve true individualized teaching.

[0053] In one embodiment, the linguistic feature analysis model is implemented through the following steps:

[0054] Transformational generative grammar theory is used to perform deep structural analysis of the target sentence and to annotate part-of-speech dependency relations and semantic roles;

[0055] Knowledge graph technology is used to calculate the semantic similarity between student expressions and standard corpora, and error type diagnostic reports are generated.

[0056] In this embodiment of the invention, the transformation-generative parsing method can identify logical breaks in complex sentence structures (inverted sentences, nested clauses) through deep structural analysis (such as topic chains and focus identification), which is superior to the limitations of traditional dependency parsing. When calculating semantic similarity, cultural context comparison is introduced (such as the difference in usage between "originate from" and "date backto") to help students understand semantic deviations caused by interference from their native language.

[0057] In one embodiment, the construction of the dynamic student profile includes:

[0058] By applying Piaget's theory of cognitive development stages to divide language ability levels, students' error patterns are classified as either symbolic deficiencies in the preoperational stage or logical fallacies in the formal operational stage.

[0059] The forgetting curve algorithm is used to predict the vocabulary memory decay cycle and automatically plan review time nodes.

[0060] In this embodiment of the invention, errors are categorized into pre-operational stage (such as concrete vocabulary association errors in young students) or formal operation stage (logical fallacies in advanced students), providing differentiated interventions, predicting vocabulary memory decay points, and automatically pushing review materials (such as news clips from the original learning context), significantly improving long-term memory retention rate.

[0061] In one embodiment, the generation of the personalized learning path includes:

[0062] For students with pronunciation deviations, a linking weakening training module is inserted, and tongue position correction animation is generated through acoustic feature comparison;

[0063] For students whose syntactic error rate is higher than the threshold, we push grammar micro-lesson videos. The video content is based on Gagné's hierarchy of learning theory, progressing step by step from concept learning to rule application.

[0064] In this embodiment of the invention, tongue position animation is generated by comparing acoustic features to visually demonstrate the difference between the student's pronunciation and the standard formant. It is especially suitable for correcting dialect accents. Based on Gagné's theory, it teaches from concept to rule progressively (such as first explaining the timeline concept of "present perfect tense" and then training practical application), avoiding cognitive overload caused by knowledge jumps.

[0065] In one embodiment, the generation of the grammar micro-lesson video further includes:

[0066] High-frequency sentence patterns are extracted from real-time news corpora, and contextualized example sentences are generated through semantic role labeling.

[0067] Based on cognitive load theory, the information density of a single video is controlled to ensure that the relevance of new knowledge points to students' existing knowledge structure exceeds a preset threshold.

[0068] In this embodiment of the invention, high-frequency sentence patterns (such as passive voice in pandemic-related reports) are extracted from the latest corpus to solve the problem of outdated textbook content, enhance learning interest, and limit the amount of information in a single video by a relevance threshold (such as new knowledge points not exceeding 30% of students' known vocabulary) to ensure efficient knowledge absorption.

[0069] In one embodiment, a teacher-side collaboration method is also included:

[0070] Automatically generate a class learning heatmap to visually present common weaknesses and the distribution of individual differences;

[0071] Group discussion topics are recommended based on the principles of communicative teaching, and the difficulty of the topics is dynamically balanced according to the cognitive labels of the members.

[0072] In this embodiment of the invention, cluster analysis is used to identify common errors in classes (such as 45% of students confusing relative pronouns in relative clauses), and targeted lesson plans are automatically recommended to reduce the burden of manual statistics on teachers. Complementary groups are matched according to cognitive styles (field-independent / field-dependent) to optimize the effectiveness of collaborative learning.

[0073] One embodiment of the present invention provides an intelligent English teaching system according to the method described in any of the above embodiments, comprising:

[0074] A linguistic feature analysis engine that integrates random forest algorithm and LSTM neural network to process multimodal learning data;

[0075] A cognitive modeling database that stores knowledge units and associated cognitive strategies organized according to Bruner's discovery learning theory.

[0076] In this embodiment of the invention, random forests process structured data (such as error rate statistics), while LSTM neural networks process unstructured data (such as the semantic coherence of essays), improving the comprehensiveness of the analysis. Knowledge points are organized according to the "discovery learning theory" (such as grammar rules presented through inquiry-based cases), promoting students' independent construction of knowledge systems.

[0077] In one embodiment, the optimization method of the linguistic feature analysis engine includes:

[0078] Adversarial training techniques are used to improve the generalization ability of dialect accents, and acoustic models are used to transfer learning to adapt to the pronunciation features of the region.

[0079] An attention mechanism is introduced to enhance the accuracy of syntax tree parsing for long and complex sentences.

[0080] In this embodiment of the invention, the system is trained by enhancing the pronunciation features of different regions (such as correcting confusion between / n / and / l / in Hunan dialect), strengthening the subject-verb consistency detection of long and difficult sentences (such as identifying subject offset in nested clauses), and improving the error localization accuracy to an industry-leading level.

[0081] In one embodiment, the method for updating the cognitive modeling database includes:

[0082] Regularly collect students' cross-platform learning behavior data and use Bayesian networks to infer the effectiveness of cognitive strategies.

[0083] Reconstruct the knowledge topology based on Ausubel's assimilation theory to ensure that the semantic distance between new concepts and existing anchor points is less than an acceptable range.

[0084] In this embodiment of the invention, the optimal cognitive strategy (such as prioritizing text and image materials for visual learners) is inferred based on cross-platform behavioral data (such as online practice and VR conversation records). When reconstructing the knowledge topology, the semantic distance between new concepts (such as "subjunctive mood") and existing anchors (such as "conditional sentences") is kept under control to avoid cognitive conflict.

[0085] One embodiment of the present invention provides a computer-readable storage medium storing program instructions, which, when executed, implement the intelligent English teaching method described above.

[0086] The device embodiments described above are merely illustrative. The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical modules; that is, they may be located in one place or distributed across multiple network modules. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0087] Through the detailed description of the above embodiments, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, including read-only memory (ROM), random access memory (RAM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), one-time programmable read-only memory (OTPROM), electrically-erasable programmable read-only memory (EEPROM), compact disc read-only memory (CD-ROM) or other optical disc storage, disk storage, magnetic tape storage, or any other computer-readable medium that can be used to carry or store data.

[0088] Finally, it should be noted that the intelligent English teaching method, system, and storage medium disclosed in the embodiments of this invention are merely preferred embodiments of the invention and are only used to illustrate the technical solutions of the invention, not to limit it. Although the invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or some of the technical features can be altered.

[0089] Line-by-line equivalent substitution; however, these modifications or substitutions do not deviate from the essence of the corresponding technical solution.

[0090] The spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. An intelligent English teaching method, characterized in that, The method comprises the following steps: A multi-dimensional linguistic feature analysis model is constructed to extract lexical features, syntactic relationship features and semantic deviation features in the student input text through natural language processing technology; A dynamic student portrait is established based on cognitive psychology theory, and the cognitive level label is updated according to real-time learning data, which includes the distribution of syntax error clusters, oral fluency indicators and vocabulary association response times; An individualized teaching path is generated, which matches teaching materials from a hierarchical resource library according to the cognitive level label, and dynamically adjusts the complexity and presentation form of the teaching strategy.

2. The intelligent English teaching method according to claim 1, characterized in that, The linguistic feature analysis model is implemented by the following steps: Deep structure analysis of the target sentence is performed using transformational grammar theory to label part-of-speech dependency relationships and semantic roles; The semantic similarity between the student's expression and the standard corpus is calculated using knowledge graph technology to generate an error type diagnosis report.

3. The intelligent English teaching method according to claim 1, characterized in that, The construction of the dynamic student portrait includes: Applying Piaget's cognitive development stage theory to divide language ability levels and categorizing student error patterns as pre-operational stage symbolization deficiencies or formal operational stage logical fallacies; Using the forgetting curve algorithm to predict vocabulary memory decay periods and automatically planning review time nodes.

4. The intelligent English teaching method according to claim 1, characterized in that, The generation of the individualized teaching path includes: For students with pronunciation deviation, a connected reading and weakened training module is inserted to generate tongue position correction animations through acoustic feature comparison; For students with a syntax error rate higher than the threshold, a grammar micro-lecture video is pushed, and the video content is based on Gagne's learning hierarchy theory and progresses from concept learning to rule application in layers.

5. The intelligent English teaching method according to claim 4, characterized in that, The generation of the grammar micro-lecture video further includes: High-frequency sentence patterns are extracted from real-time news corpus, and scenario-based example sentences are generated through semantic role labeling; based on the cognitive load theory, the information density of a single video is controlled to ensure that the relevance of new knowledge points to the student's existing knowledge structure exceeds the preset threshold.

6. The intelligent English teaching method according to claim 1, characterized in that, It also includes a teacher-side collaboration method: A class learning situation heat map is automatically generated to visually present common weaknesses and individual difference distribution; based on the principles of communicative language teaching, group discussion topics are recommended, and the difficulty of the topics is dynamically balanced according to the cognitive labels of the group members.

7. An intelligent English teaching system according to any one of the methods of claims 1-6, characterized in that, It includes: A linguistic feature analysis engine that integrates random forest algorithm and LSTM neural network to process multi-modal learning data; A cognitive modeling database that stores knowledge units and associated cognitive strategies organized according to Bruner's discovery learning theory.

8. The system of claim 7, wherein, The optimization method of the linguistic feature analysis engine includes: Using adversarial training technology to improve dialect accent generalization ability, and through acoustic model transfer learning to adapt to regional pronunciation features; Introducing attention mechanism to enhance the accuracy of syntax tree analysis for long and difficult sentences.

9. The system of claim 7, wherein, The updating method of the cognitive modeling database includes: Regularly collecting student cross-platform learning behavior data and inferring the effectiveness of cognitive strategies through Bayesian networks; According to O'Sullivan's assimilation theory, the knowledge topology is reconstructed to ensure that the semantic distance between new concepts and existing fixed points is less than the acceptable range.

10. A computer readable storage medium storing program instructions, characterized in that, The instructions, when executed, implement the method of any one of claims 1-9.

Images