Automatic textbook construction and optimization method and device

Abstract

The invention relates to the technical field of textbook construction and optimization, in particular to an automatic textbook construction and optimization method and device.The method comprises the following steps that original resources of multiple types of textbooks are collected, and core knowledge elements in the original resources are recognized; constructing a structured mapping relationship between the knowledge elements and the teaching resources based on the core knowledge elements; on the basis of the structured mapping relationship, constructing a knowledge graph covering multidisciplinary knowledge points and a preposition dependency relationship, and automatically identifying knowledge faults and complementing missing association so as to form a knowledge network with self-evolution ability; based on the knowledge network, utilizing a hybrid generation model to cooperatively work, and quickly generating a structured textbook framework containing knowledge veins, teaching cases and exercise resources; learning condition data including classroom behaviors, learning feedback and knowledge mastering conditions are obtained, and a quantity and chemical condition report is obtained after data processing; and updating data based on the quantitative chemical situation report and the knowledge graph.

Description

Technical Field

[0001] This invention relates to the field of textbook construction and optimization technology, and in particular to an automated method and apparatus for textbook construction and optimization. Background Technology

[0002] Traditional textbook development and optimization is a process that heavily relies on human experience, and it suffers from the following problems: Traditional paper textbooks or simple electronic textbooks are difficult to change once published, making it impossible to keep up with the rapid pace of knowledge iteration and to personalize them for different student groups.

[0003] Teaching resources (such as videos, exercises, and case studies) often exist in isolation and have a weak connection with the core knowledge points of the textbook. Teachers need to spend a lot of time manually searching for and integrating resources, resulting in low lesson preparation efficiency.

[0004] High-quality lesson plans and teaching strategies rely heavily on the personal experience of excellent teachers, making them difficult to distill, replicate, and scale up, resulting in uneven teaching quality.

[0005] The evaluation of teaching effectiveness relies heavily on periodic exams, which result in delayed feedback. Teachers are unable to grasp the learning situation of the class and individuals in real time during the teaching process, making it difficult to adjust teaching content and strategies in a timely and accurate manner.

[0006] Most existing knowledge graphs are statically constructed and lack the ability to self-optimize and evolve based on actual teaching feedback, thus failing to form a virtuous cycle of becoming smarter with use.

[0007] In recent years, although tools that use artificial intelligence technology to assist education have emerged, such as intelligent question banks and resource recommendation systems, these tools are mostly fragmented solutions targeting single links. They have failed to fundamentally connect the entire chain from resource processing, knowledge construction, content generation, learning analysis to dynamic optimization, and have failed to form a closed-loop education ecosystem with self-evolution capabilities. Summary of the Invention

[0008] Therefore, it is necessary to provide an automated method and device for constructing and optimizing teaching materials to address the aforementioned technical problems. This fundamentally changes the static and generic model of traditional teaching materials, enabling them to be personalized, dynamic, and intelligent.

[0009] This invention provides an automated method for constructing and optimizing teaching materials, the method comprising: Collect original resources from multiple types of teaching materials and identify the core knowledge elements in the original resources; Construct a structured mapping relationship between knowledge elements and teaching resources based on core knowledge elements; Based on the structured mapping relationship, a knowledge graph covering knowledge points from multiple disciplines and prerequisite dependencies is constructed, and knowledge gaps are automatically identified and missing connections are filled in, thereby forming a knowledge network with self-evolution capabilities. Based on the knowledge network, a structured textbook framework containing knowledge context, teaching cases, and exercise resources is quickly generated by using a hybrid generative model to work collaboratively. Acquire learning data including classroom behavior, learning feedback, and knowledge mastery, and obtain a quantitative learning report after data processing; Based on the quantitative learning reports and the updated data of the knowledge graph, the content structure, resource allocation, and presentation format of the teaching materials are automatically adjusted.

[0010] In one embodiment, the process of collecting original resources from multiple types of teaching materials and identifying core knowledge elements within those resources includes: Text, charts, and formulas were identified as the core data collection types. Resources were collected through authorized acquisition, teacher uploads, and compliant crawling. Irrelevant information was removed and the format was standardized to form a well-organized collection of original textbook resources. For the original set of teaching materials, key concepts and definitions in the text are extracted based on semantic understanding, annotation information in the charts is identified and associated with themes through scene text detection, and mathematical expressions are parsed and associated with knowledge points using a formula recognizer. The extracted and identified content is screened, classified, and organized to form a list of core knowledge elements.

[0011] In one embodiment, the construction of a structured mapping relationship between knowledge elements and teaching resources based on core knowledge elements includes: Based on the aforementioned list of core knowledge elements, teaching resources, including videos, animations, and exercises, are manually reviewed and intelligently annotated to establish a precise correspondence between knowledge points and resources. The labeled correspondences are structured and encoded according to preset fields, and stored in a relational database to construct a mapping table that can be queried efficiently.

[0012] In one embodiment, the construction of a knowledge graph covering multidisciplinary knowledge points and prerequisite dependencies based on the structured mapping relationship, and the automatic identification of knowledge gaps and completion of missing connections, thereby forming a knowledge network with self-evolutionary capabilities, includes: Based on the subject standards, entity types and relationship types are defined. The elements in the core knowledge element list and their relationships defined through the structured mapping relationship are imported into the graph database to construct a preliminary knowledge graph. Based on the aforementioned knowledge graph prototype, the graph algorithm is used to infer potential dependencies between entities and to identify sparsely connected regions as knowledge gaps. Based on rules or correlation, new associations are automatically created for the knowledge gaps, enabling the dynamic expansion and self-evolution of the knowledge network.

[0013] In one embodiment, the step of rapidly generating a structured textbook framework containing knowledge context, teaching cases, and exercise resources based on the knowledge network and utilizing a hybrid generative model includes: Using the knowledge points in the knowledge network as the framework, the rapid generation model is called to fill in the teaching cases and exercises, and the first draft of the lesson plan is output. The initial draft of the lesson plan is fed into the logic optimization model to verify the accuracy of knowledge and optimize the teaching process. It is then reorganized according to the standard template to generate the structured teaching material framework.

[0014] In one embodiment, the step of acquiring learning data including classroom behavior, learning feedback, and knowledge mastery, and processing the data to obtain a quantitative learning report, includes: Classroom cameras are used to collect videos of students' behavior, and answer data is simultaneously obtained from the learning system. The collected raw data is cleaned, denoised, and timestamp aligned to form a well-organized multimodal learning dataset; The multimodal learning dataset is analyzed to calculate quantitative indicators including head-up rate, number of interactions, and test accuracy. After comprehensive evaluation, a visual learning report containing data charts and conclusions is generated.

[0015] In one embodiment, automatically adjusting the textbook content structure, resource allocation, and presentation format based on the updated data from the quantitative learning report and the knowledge graph includes: Analyze the visualized learning report to pinpoint common knowledge weaknesses in the class; Based on the identified knowledge gaps, targeted explanation videos and practice questions are selected and pushed from teaching resources associated with the knowledge graph, and suggestions for teachers to adjust their teaching focus are provided. To address individual learning differences among students, the system dynamically adjusts the order of their learning paths within the knowledge graph and triggers content generation services to customize personalized learning materials, including review outlines or explanations of incorrect questions, based on their current knowledge status.

[0016] The present invention also provides an automated textbook construction and optimization device, the device comprising: The multi-source knowledge extraction module is used to collect original resources from multiple types of teaching materials and identify the core knowledge elements in the original resources. The resource intelligent association module is used to construct a structured mapping relationship between knowledge elements and teaching resources based on core knowledge elements; The graph self-evolution module is used to construct a knowledge graph covering knowledge points and prerequisite dependencies from multiple disciplines based on the structured mapping relationship, and automatically identify knowledge gaps and fill in missing connections, thereby forming a knowledge network with self-evolution capabilities. The lesson plan generation module is used to quickly generate a structured teaching material framework containing knowledge context, teaching cases, and exercise resources based on the knowledge network and by working collaboratively with a hybrid generation model. The learning analysis module is used to acquire learning data including classroom behavior, learning feedback, and knowledge mastery, and to generate a quantitative learning report after data processing. The dynamic optimization module is used to automatically adjust the content structure, resource allocation, and presentation format of the teaching materials based on the updated data of the quantitative learning report and the knowledge graph.

[0017] The present invention also provides an electronic device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the automated teaching material construction and optimization method as described above.

[0018] The present invention also provides a computer storage medium storing a computer program, which, when executed by a processor, implements the automated textbook construction and optimization method as described above.

[0019] The present invention also provides a computer program product, including a computer program that, when executed by a processor, implements the automated textbook construction and optimization method as described above.

[0020] The aforementioned automated textbook construction and optimization method and device constructs a complete closed-loop system from perception, decision-making, execution to feedback. It automatically extracts knowledge from multimodal resources, constructs a dynamically growing knowledge network, and intelligently generates teaching content based on this network. Finally, it drives the adaptive optimization of the entire system through real learning data. This method fundamentally changes the static and general model of traditional textbooks, realizing the personalization, dynamism and intelligence of textbooks. Attached Figure Description

[0021] To more clearly illustrate the technical solutions in this 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 some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0022] Figure 1 The flowchart of the automated teaching material construction and optimization method provided by this invention; Figure 2Module diagram of the automated teaching material construction and optimization device provided by the present invention Figure 3 An internal structural diagram of the computer device provided by the present invention. Detailed Implementation

[0023] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, 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.

[0024] The following is combined with Figures 1 to 3 This invention describes an automated method and apparatus for constructing and optimizing teaching materials.

[0025] In one embodiment, an automated teaching material construction and optimization method includes the following steps: Collect original resources from multiple types of teaching materials and identify the core knowledge elements in the original resources; Construct a structured mapping relationship between knowledge elements and teaching resources based on core knowledge elements; Based on the structured mapping relationship, a knowledge graph covering knowledge points from multiple disciplines and prerequisite dependencies is constructed, and knowledge gaps are automatically identified and missing connections are filled in, thereby forming a knowledge network with self-evolution capabilities. Based on the knowledge network, a structured textbook framework containing knowledge context, teaching cases, and exercise resources is quickly generated by using a hybrid generative model to work collaboratively. Acquire learning data including classroom behavior, learning feedback, and knowledge mastery, and obtain a quantitative learning report after data processing; Based on the quantitative learning reports and the updated data of the knowledge graph, the content structure, resource allocation, and presentation format of the teaching materials are automatically adjusted.

[0026] The aforementioned automated textbook construction and optimization method automatically collects original resources from various types of textbooks and identifies core knowledge elements. Based on these core knowledge elements, it constructs a structured mapping relationship between them and teaching resources such as videos and exercises. Based on this structured mapping relationship, it builds a knowledge graph covering mechanics knowledge points and their prerequisite dependencies. It can also automatically identify knowledge gaps, fill in missing connections, and form a knowledge network with self-evolutionary capabilities. Based on this knowledge network, it uses a hybrid generative model to work collaboratively to quickly generate a structured textbook framework containing knowledge context, teaching cases, and exercises. Then, during the teaching process, it acquires learning data such as students' classroom behavior, learning feedback, and knowledge mastery, processes it, and obtains a quantitative learning report. Finally, based on the quantitative learning report and the updated data of the knowledge graph, it automatically adjusts the content structure, resource allocation, and presentation format of the textbook. This method constructs a complete closed-loop system encompassing perception, decision-making, execution, and feedback. It automatically extracts knowledge from multimodal resources, builds a dynamically growing knowledge network, and intelligently generates teaching content based on this network. Finally, it drives the adaptive optimization of the entire system through real learning data. This method fundamentally changes the static and generic model of traditional textbooks, realizing the personalization, dynamism, and intelligence of textbooks.

[0027] In one embodiment, the process of collecting original resources from multiple types of teaching materials and identifying core knowledge elements within those resources includes the following steps: Text, charts, and formulas were identified as the core data collection types. Resources were collected through authorized acquisition, teacher uploads, and compliant crawling. Irrelevant information was removed and the format was standardized to form a well-organized collection of original textbook resources. Specifically, a diversified resource input channel is built through three methods: "authorized acquisition" (such as cooperation with publishers), "teacher uploads" (aggregating personalized resources), and "compliant crawling" (acquiring from open source platforms). "Removing irrelevant information" (such as advertisements and irrelevant watermarks) and "unifying formats" (such as unifying images to RGB mode and text to UTF-8 encoding) are key to data preprocessing, directly improving the accuracy and efficiency of subsequent recognition steps.

[0028] For the original set of teaching materials, key concepts and definitions in the text are extracted based on semantic understanding, annotation information in the charts is identified and associated with themes through scene text detection, and mathematical expressions are parsed and associated with knowledge points using a formula recognizer. Specifically, "semantic understanding-based" refers to using natural language processing technologies, such as named entity recognition, to locate and extract key concepts and their definitions from text, such as "Newton's First Law" and "trigonometric functions." "Scene text detection" refers to using computer vision technology to identify text such as titles, legends, and axis labels in charts and graphs, and to associate them with the main content of the chart to understand its expressed "theme," such as identifying a line chart as a "temperature change chart for the past ten years." "Formula recognizer" is specifically designed for subjects such as mathematics and physics, converting formulas in images into structured, computable formats such as LaTeX, and associating them with knowledge points such as "calculus" and "Ohm's Law."

[0029] The extracted and identified content is screened, classified, and organized to form a list of core knowledge elements.

[0030] Specifically, by setting rules (such as removing words that appear too infrequently) or using classification models, the extracted elements are organized into categories such as "concepts", "theorems", "experiments", and "people", ultimately forming a clear and well-categorized "list of core knowledge elements".

[0031] In one embodiment, constructing a structured mapping relationship between knowledge elements and teaching resources based on core knowledge elements includes the following steps: Based on the aforementioned list of core knowledge elements, teaching resources, including videos, animations, and exercises, are manually reviewed and intelligently annotated to establish a precise correspondence between knowledge points and resources. Specifically, the NLP model is first used to analyze the content of resources (such as video subtitles and exercise questions) and automatically recommend the knowledge points that may be related to them. Then, the teacher performs "manual review" and confirmation, or directly creates new associations.

[0032] The labeled correspondences are structured and encoded according to preset fields, and stored in a relational database to construct a mapping table that can be queried efficiently.

[0033] Specifically, "preset fields" can include knowledge point ID, resource ID, resource type (video / exercise / animation), correlation strength (core / extension), applicable grade level, etc. Storing this information in a "relational database" (such as MySQL) creates a powerful "mapping table," enabling the system to quickly find all related resources for a specific knowledge point, or all knowledge points related to a specific resource, through SQL queries.

[0034] In one embodiment, the construction of a knowledge graph covering multidisciplinary knowledge points and prerequisite dependencies based on the structured mapping relationship, and the automatic identification of knowledge gaps and the filling in of missing connections to form a self-evolving knowledge network, includes the following steps: Based on the subject standards, entity types and relationship types are defined. The elements in the core knowledge element list and their relationships defined through the structured mapping relationship are imported into the graph database to construct a preliminary knowledge graph. For example, define entity types as concepts, chapters, and characters, and relation types as contained in, preceded by, and related to.

[0035] Then, the "core knowledge element list" is used as an entity, and the explicit associations in the "mapping relationship" (such as "Chapter A contains knowledge point B") are used as initial relations. These are then imported into a "graph database" (such as Neo4j) to form a "prototype of knowledge graph".

[0036] Based on the aforementioned knowledge graph prototype, the graph algorithm is used to infer potential dependencies between entities and to identify sparsely connected regions as knowledge gaps. Specifically, by running graph algorithms such as graph path analysis and community discovery, potential relationships that are not explicitly labeled can be inferred. For example, it can be found that "knowledge point C" and "knowledge point D" are frequently queried by the same group of students, thus inferring that they have a potential connection.

[0037] At the same time, the algorithm can identify regions in the network with very sparse connections, which are called "knowledge gaps" and represent the imperfections of the current knowledge system.

[0038] Based on rules or correlation, new associations are automatically created for the knowledge gaps, enabling the dynamic expansion and self-evolution of the knowledge network.

[0039] Specifically, "rule-based" can mean: if two knowledge points belong to the same larger parent knowledge point, a "related" relationship is automatically established. "Relevance-based" can mean: when the co-occurrence frequency or semantic similarity of two knowledge points exceeds a threshold, an association is automatically created. In this way, the knowledge network can achieve "dynamic expansion and self-evolution".

[0040] In one embodiment, the step of rapidly generating a structured textbook framework containing knowledge context, teaching cases, and exercise resources based on the knowledge network and utilizing a hybrid generative model includes the following steps: Using the knowledge points in the knowledge network as the framework, the rapid generation model is called to fill in the teaching cases and exercises, and the first draft of the lesson plan is output. Specifically, the "knowledge point framework" originates from the "knowledge network," providing the logical framework of the lesson plan. Based on this framework, a "rapid generation model" (such as DiffusionLM) generates draft content for teaching cases and exercises in parallel and rapidly, quickly outputting a comprehensive "first draft of the lesson plan." The goal at this stage is to quickly generate a large amount of alternative content.

[0041] The initial draft of the lesson plan is fed into the logic optimization model to verify the accuracy of knowledge and optimize the teaching process. It is then reorganized according to the standard template to generate the structured teaching material framework.

[0042] Specifically, "logic optimization models" (such as the CausalLM / QwQ model) are used to further refine the initial draft, including: 1) "Knowledge Accuracy Verification": Verify whether the generated factual content is correct; 2) "Teaching process optimization": Adjust the order of teaching steps to better align with cognitive patterns; 3) "Reorganize according to standard templates": Fill the content into standard modules such as "teaching objectives, key points and difficulties, teaching process and homework", and finally output a "structured textbook framework" that can be used directly.

[0043] In one embodiment, the step of acquiring learning data including classroom behavior, learning feedback, and knowledge mastery, and processing the data to obtain a quantitative learning report, includes the following steps: Classroom cameras are used to collect videos of students' behavior, and answer data is simultaneously obtained from the learning system. Specifically, "behavioral videos" record students' nonverbal feedback, while "answer data" reflects their knowledge mastery. The combination of the two forms the data foundation for "multimodal" learning.

[0044] The collected raw data is cleaned, denoised, and timestamp aligned to form a well-organized multimodal learning dataset; Specifically, "cleaning and noise reduction" includes filtering out blurry video frames and correcting incorrect answer records. "Timestamp alignment" is crucial, as it ensures that the behavior of "students looking down at the 10th minute" can be accurately associated with "the knowledge points explained at the 10th minute," thereby forming a "well-organized multimodal learning dataset."

[0045] The multimodal learning dataset is analyzed to calculate quantitative indicators including head-up rate, number of interactions, and test accuracy. After comprehensive evaluation, a visual learning report containing data charts and conclusions is generated.

[0046] Specifically, computer vision algorithms are used to calculate "head-up rate," speech recognition and text analysis are used to count "interaction frequency," and the grading system is used to calculate "test accuracy." These indicators are comprehensively evaluated (e.g., a weighted "classroom focus index"), and a "visualized learning report" with charts (such as trend graphs and heat maps) is generated, making the learning situation clear at a glance.

[0047] In one embodiment, automatically adjusting the content structure, resource allocation, and presentation format of teaching materials based on the updated data from the quantitative learning report and the knowledge graph includes the following steps: Analyze the visualized learning report to pinpoint common knowledge weaknesses in the class; For example, if it is found that the accuracy rate of a class in solving problems related to "trigonometric function reduction formulas" is generally below 60%, it can be identified as a "common weak point in the class's knowledge".

[0048] Based on the identified knowledge gaps, targeted explanation videos and practice questions are selected and pushed from teaching resources associated with the knowledge graph, and suggestions for teachers to adjust their teaching focus are provided. Specifically, based on weak knowledge points, the system automatically selects targeted explanation videos and practice questions from the "mapping relationship table" and pushes them to all students in the class. At the same time, it generates "teaching focus adjustment suggestions" for teachers, such as "It is recommended to review the trigonometric function reduction formulas in the next lesson."

[0049] To address individual learning differences among students, the system dynamically adjusts the order of their learning paths within the knowledge graph and triggers content generation services to customize personalized learning materials, including review outlines or explanations of incorrect questions, based on their current knowledge status.

[0050] For example, for a student who is weak in "Knowledge Point A", the system will "dynamically adjust" their "learning path order" in the "knowledge graph", automatically inserting a review session for "Knowledge Point A" before they learn "Knowledge Point B". It will also "trigger content generation service" to "customize" a "personalized learning material" containing only the questions they answered incorrectly.

[0051] The automated textbook construction and optimization apparatus provided by the present invention is described below. The automated textbook construction and optimization apparatus described below can be referred to in correspondence with the automated textbook construction and optimization method described above.

[0052] In one embodiment, an automated teaching material construction and optimization device includes a multi-source knowledge extraction module, a resource intelligent association module, a graph self-evolution module, a lesson plan generation module, a learning situation analysis module, and a dynamic optimization module. The multi-source knowledge extraction module is used to collect original resources from multiple types of teaching materials and identify the core knowledge elements in the original resources; The resource intelligent association module is used to construct a structured mapping relationship between knowledge elements and teaching resources based on core knowledge elements; The graph self-evolution module is used to construct a knowledge graph covering knowledge points and prerequisite dependencies from multiple disciplines based on the structured mapping relationship, and automatically identify knowledge gaps and fill in missing connections, thereby forming a knowledge network with self-evolution capabilities. The lesson plan generation module is used to quickly generate a structured teaching material framework containing knowledge context, teaching cases, and exercise resources based on the knowledge network and by working collaboratively with a hybrid generation model. The learning analysis module is used to acquire learning data including classroom behavior, learning feedback, and knowledge mastery, and to generate a quantitative learning report after data processing. The dynamic optimization module is used to automatically adjust the content structure, resource configuration, and presentation format of the teaching materials based on the updated data of the quantitative learning report and the knowledge graph.

[0053] The aforementioned automated textbook construction and optimization device constructs a complete closed-loop system encompassing perception, decision-making, execution, and feedback. It automatically extracts knowledge from multimodal resources, builds a dynamically growing knowledge network, and intelligently generates teaching content based on this network. Finally, it drives the adaptive optimization of the entire system through real learning data. This method fundamentally changes the static and generic model of traditional textbooks, realizing the personalization, dynamism, and intelligence of textbooks.

[0054] Figure 3 This example illustrates a schematic diagram of the physical structure of an electronic device, which can be a smart terminal. Its internal structure diagram can be as follows: Figure 3 As shown, this electronic device includes a processor, memory, and a network interface connected via a system bus. The processor provides computing and control capabilities. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The network interface is used to communicate with external terminals via a network connection. When the computer program is executed by the processor, it implements an automated teaching material construction and optimization method.

[0055] Those skilled in the art will understand that Figure 3 The structure shown is merely a block diagram of a portion of the structure related to the present invention and does not constitute a limitation on the electronic device to which the present invention is applied. A specific electronic device may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0056] On the other hand, the present invention also provides a computer storage medium storing a computer program, which, when executed by a processor, implements an automated textbook construction and optimization method.

[0057] On another front, a computer program product or computer program is provided, comprising computer instructions stored in a computer storage medium. The processor of an electronic device reads the computer instructions from the computer storage medium, and when the processor executes the computer instructions, it implements an automated teaching material construction and optimization method.

[0058] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. This computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided by this invention can include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory may include random access memory (RAM) or external cache memory.

[0059] By way of illustration and not limitation, RAM is available in a variety of forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0060] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0061] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these modifications and improvements all fall within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the appended claims.

Claims

1. A method for constructing and optimizing automated teaching materials, characterized in that, The method includes: Collect original resources from multiple types of teaching materials and identify the core knowledge elements in the original resources; Construct a structured mapping relationship between knowledge elements and teaching resources based on core knowledge elements; Based on the structured mapping relationship, a knowledge graph covering knowledge points from multiple disciplines and prerequisite dependencies is constructed, and knowledge gaps are automatically identified and missing connections are filled in, thereby forming a knowledge network with self-evolution capabilities. Based on the knowledge network, a structured textbook framework containing knowledge context, teaching cases, and exercise resources is quickly generated by using a hybrid generative model to work collaboratively. Acquire learning data including classroom behavior, learning feedback, and knowledge mastery, and obtain a quantitative learning report after data processing; Based on the quantitative learning reports and the updated data of the knowledge graph, the content structure, resource allocation, and presentation format of the teaching materials are automatically adjusted.

2. The automated textbook construction and optimization method according to claim 1, characterized in that, The process involves collecting original resources from multiple types of teaching materials and identifying the core knowledge elements within these resources, including: Text, charts, and formulas were identified as the core data collection types. Resources were collected through authorized acquisition, teacher uploads, and compliant crawling. Irrelevant information was removed and the format was standardized to form a well-organized collection of original textbook resources. For the original set of teaching materials, key concepts and definitions in the text are extracted based on semantic understanding, annotation information in the charts is identified and associated with themes through scene text detection, and mathematical expressions are parsed and associated with knowledge points using a formula recognizer. The extracted and identified content is screened, classified, and organized to form a list of core knowledge elements.

3. The automated textbook construction and optimization method according to claim 2, characterized in that, The structured mapping relationship between knowledge elements and teaching resources based on core knowledge elements includes: Based on the aforementioned list of core knowledge elements, teaching resources, including videos, animations, and exercises, are manually reviewed and intelligently annotated to establish a precise correspondence between knowledge points and resources. The labeled correspondences are structured and encoded according to preset fields, and stored in a relational database to construct a mapping table that can be queried efficiently.

4. The automated textbook construction and optimization method according to claim 3, characterized in that, Based on the structured mapping relationship, a knowledge graph covering multidisciplinary knowledge points and prerequisite dependencies is constructed, and knowledge gaps are automatically identified and missing connections are filled in, thereby forming a knowledge network with self-evolution capabilities, including: Based on the subject standards, entity types and relationship types are defined. The elements in the core knowledge element list and their relationships defined through the structured mapping relationship are imported into the graph database to construct a preliminary knowledge graph. Based on the aforementioned knowledge graph prototype, the graph algorithm is used to infer potential dependencies between entities and to identify sparsely connected regions as knowledge gaps. Based on rules or correlation, new associations are automatically created for the knowledge gaps, enabling the dynamic expansion and self-evolution of the knowledge network.

5. The automated textbook construction and optimization method according to claim 4, characterized in that, Based on the knowledge network, a hybrid generative model is used to collaboratively generate a structured textbook framework that includes knowledge context, teaching cases, and exercise resources, including: Using the knowledge points in the knowledge network as the framework, the rapid generation model is called to fill in the teaching cases and exercises, and the first draft of the lesson plan is output. The initial draft of the lesson plan is fed into the logic optimization model to verify the accuracy of knowledge and optimize the teaching process. It is then reorganized according to the standard template to generate the structured teaching material framework.

6. The automated textbook construction and optimization method according to claim 5, characterized in that, The process of acquiring learning data including classroom behavior, learning feedback, and knowledge mastery, and then processing the data to obtain a quantitative learning report, includes: Classroom cameras are used to collect videos of students' behavior, and answer data is simultaneously obtained from the learning system. The collected raw data is cleaned, denoised, and timestamp aligned to form a well-organized multimodal learning dataset; The multimodal learning dataset is analyzed to calculate quantitative indicators including head-up rate, number of interactions, and test accuracy. After comprehensive evaluation, a visual learning report containing data charts and conclusions is generated.

7. The automated textbook construction and optimization method according to claim 6, characterized in that, The automatic adjustment of the textbook content structure, resource allocation, and presentation format based on the updated data from the quantitative learning report and the knowledge graph includes: Analyze the visualized learning report to pinpoint common knowledge weaknesses in the class; Based on the identified knowledge gaps, targeted explanation videos and practice questions are selected and pushed from teaching resources associated with the knowledge graph, and suggestions for teachers to adjust their teaching focus are provided. To address individual learning differences among students, the system dynamically adjusts the order of their learning paths within the knowledge graph and triggers content generation services to customize personalized learning materials, including review outlines or explanations of incorrect questions, based on their current knowledge status.

8. An automated textbook construction and optimization device, characterized in that, The device includes: The multi-source knowledge extraction module is used to collect original resources from multiple types of teaching materials and identify the core knowledge elements in the original resources. The resource intelligent association module is used to construct a structured mapping relationship between knowledge elements and teaching resources based on core knowledge elements; The graph self-evolution module is used to construct a knowledge graph covering knowledge points and prerequisite dependencies from multiple disciplines based on the structured mapping relationship, and automatically identify knowledge gaps and fill in missing connections, thereby forming a knowledge network with self-evolution capabilities. The lesson plan generation module is used to quickly generate a structured teaching material framework containing knowledge context, teaching cases, and exercise resources based on the knowledge network and by working collaboratively with a hybrid generation model. The learning analysis module is used to acquire learning data including classroom behavior, learning feedback, and knowledge mastery, and to generate a quantitative learning report after data processing. The dynamic optimization module is used to automatically adjust the content structure, resource allocation, and presentation format of the teaching materials based on the updated data of the quantitative learning report and the knowledge graph.

9. An electronic device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 7.

10. A computer storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 7.

Images