Teaching material data generation method and device, equipment, medium and program product

Abstract

The invention discloses a textbook data generation method and device, equipment, a medium and a program product. Generating initial teaching material data based on the teaching audio and video and the matched teaching plan data; acquiring learning behavior data and homework data of the students on the teaching audio and video, analyzing the learning behavior data and the homework data through a knowledge point evaluation model constructed based on a large language model (LLM), and outputting a mastery degree evaluation result of the students on knowledge points; and inputting the mastery degree evaluation result into a strategy optimization model constructed based on LLM, outputting at least one dimension of adjustment parameters through the strategy optimization model, generating a textbook optimization strategy according to the adjustment parameters, and optimizing the initial textbook data based on the textbook optimization strategy. According to the scheme, the problems that existing teaching material content is static and cannot be updated, personalized customization is lacked, and a feedback evaluation mechanism is lacked can be solved.

Description

Technical Field

[0001] This invention relates to the field of digital education technology, and in particular to a method, apparatus, equipment, medium, and program product for generating teaching material data. Background Technology

[0002] With the development of the Internet, online education and blended learning models have become increasingly popular. Students' learning needs have become more diverse than ever before. Different learners have significant differences in learning speed, knowledge background, and interest preferences, and the demand for personalized education and precise teaching is becoming stronger.

[0003] Current teaching still uses standardized teaching materials, namely paper textbooks and e-books in specific formats. However, the content of standardized teaching materials is relatively fixed, making it impossible to adjust the content according to individual student differences, and it is also difficult to provide immediate and in-depth learning feedback, resulting in low learning efficiency and a poor learning experience.

[0004] Therefore, a textbook generation method is needed to address the problems of existing textbooks being static and unable to be updated, lacking personalized customization, and lacking feedback and evaluation mechanisms, thus failing to meet the requirements of education and teaching for personalized and dynamic content. Summary of the Invention

[0005] This invention provides a method, apparatus, device, medium, and program product for generating textbook data, in order to solve the problems of existing textbook content being static and unable to be updated, lacking personalized customization, and lacking feedback and evaluation mechanisms.

[0006] According to one aspect of the present invention, a method for generating textbook data is provided, comprising:

[0007] Initial teaching material data is generated based on the lecture audio and video and accompanying lesson plan data;

[0008] The system collects students' learning behavior data and homework data related to the lecture audio and video. It then analyzes the learning behavior data and homework data using a knowledge point assessment model built on the Large Language Model (LLM) to output an assessment result of the students' mastery of the knowledge points. The learning behavior data includes at least one of the following: the distribution of time spent watching the lecture audio and video, the time points when notes were taken, the frequency and length of notes, and interaction data with the course content.

[0009] The homework data shall include at least one of the following: homework completion time, homework accuracy rate, and question difficulty level;

[0010] The mastery assessment results are input into a strategy optimization model based on LLM. The strategy optimization model outputs adjustment parameters for at least one dimension. A textbook optimization strategy is generated based on the adjustment parameters. The initial textbook data is then optimized based on the textbook optimization strategy.

[0011] Optionally, the generation of initial teaching material data based on the lecture audio / video and accompanying lesson plan data includes:

[0012] The speech recognition tool is used to convert the lecture audio and video into lecture text with timestamps, and the content of the lecture text corresponds to the specific playback time of the lecture audio and video;

[0013] The teaching text and the accompanying lesson plan data are input into the knowledge point assessment model, and the extracted core classroom teaching information is output; the logical relationship between each knowledge point in the core teaching information is determined; and a teaching syllabus framework is generated based on the core teaching information and the logical relationship.

[0014] Using the teaching syllabus framework as input, an intelligent agent is generated by calling the content. The chapter structure of the teaching materials is planned according to the teaching syllabus framework, and the length ratio of each chapter is determined. Based on the length ratio, the core teaching information is transformed into teaching material text.

[0015] The initial teaching material is obtained by selecting practice questions corresponding to the knowledge points of each chapter from a pre-set question bank and using the textbook text and the practice questions.

[0016] Optionally, the question bank includes multiple question types, and each question type includes basic practice questions and advanced practice questions divided according to difficulty thresholds; correspondingly, determining the practice questions corresponding to each chapter's knowledge points from the preset question bank includes:

[0017] Determine the fit between the knowledge points of each chapter and each question type, and determine the first proportion of each question type in each chapter based on the ratio of the fit between the knowledge points of each chapter and each question type;

[0018] Based on the first proportion and the preset second proportion of basic and advanced practice questions, practice questions corresponding to the knowledge points of each chapter are determined from the preset question bank.

[0019] Optionally, after obtaining the teaching material text, the method further includes:

[0020] The semantic validation of the teaching material text is performed using LLM to determine whether the teaching material text conforms to the preset teaching material specifications and whether the text content of the teaching material text in each chapter matches the core teaching information.

[0021] Optionally, the analysis of the learning behavior data and homework data using a knowledge point assessment model built on LLM, and the output of assessment results on students' mastery of different knowledge points, includes:

[0022] The knowledge point assessment model is used to weight and evaluate the dwell time distribution, the time point of note-taking, the note frequency, the note content length, and the interaction behavior data to obtain a preliminary mastery assessment result.

[0023] The preliminary mastery assessment results are verified and corrected based on the ratio of the completion time of the assignment to the average completion time of the assignment, and the ratio of the accuracy rate of the assignment to the average accuracy rate of the assignment, under different difficulty levels, to obtain the mastery assessment results.

[0024] Optionally, the step of outputting adjustment parameters of at least one dimension through the strategy optimization model, and generating a textbook optimization strategy based on the adjustment parameters, includes:

[0025] For knowledge points assessed as weak in the mastery assessment results, the content weight adjustment parameters, explanation depth adjustment parameters, and length ratio adjustment parameters for that knowledge point are increased, and the number of basic practice questions is increased.

[0026] For knowledge points assessed as mastered or proficient in the mastery assessment results, the content weight adjustment parameters, explanation depth adjustment parameters, and length ratio adjustment parameters for that knowledge point are reduced, while the number of advanced practice questions is increased.

[0027] According to another aspect of the present invention, a textbook data generation apparatus is provided, comprising:

[0028] The initial teaching material data generation unit is used to generate initial teaching material data based on the lecture audio and video and the accompanying lesson plan data;

[0029] The evaluation result output unit is used to collect students' learning behavior data and homework data on the teaching audio and video, analyze the learning behavior data and homework data through a knowledge point evaluation model built based on the Large Language Model (LLM), and output the evaluation results of students' mastery of knowledge points.

[0030] The initial textbook optimization unit is used to input the mastery assessment results into a strategy optimization model based on LLM, output adjustment parameters of at least one dimension through the strategy optimization model, generate a textbook optimization strategy based on the adjustment parameters, and optimize the initial textbook data based on the textbook optimization strategy.

[0031] According to another aspect of the present invention, an electronic device is provided, the electronic device comprising:

[0032] At least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores a computer program executable by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the textbook data generation method according to any embodiment of the present invention.

[0033] According to another aspect of the present invention, a computer-readable storage medium is provided, the computer-readable storage medium storing computer instructions, the computer instructions being configured to cause a processor to execute and implement the textbook data generation method according to any embodiment of the present invention.

[0034] According to another aspect of the present invention, a computer program product is provided, the computer program product comprising a computer program that, when executed by a processor, implements the textbook data generation method described in any embodiment of the present invention.

[0035] The technical solution of this invention realizes the automated generation of teaching materials, greatly improves production efficiency, and ensures that the content matches the teaching audio and video and the accompanying lesson plan data. Based on the accurate evaluation of multi-dimensional learning behavior data and homework data, combined with LLM to dynamically adjust the teaching materials, it realizes personalized adaptation to the learning needs of different students, and solves the pain points of traditional teaching materials being static and fixed, lacking personalization and feedback mechanisms.

[0036] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of the present invention, nor is it intended to limit the scope of the invention. Other features of the invention will become readily apparent from the following description. Attached Figure Description

[0037] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0038] Figure 1 This is a flowchart of a textbook data generation method provided in an embodiment of the present invention;

[0039] Figure 2 This is a flowchart of an initial teaching material data generation method provided in an embodiment of the present invention;

[0040] Figure 3 This is a flowchart of a method for outputting mastery assessment results provided in an embodiment of the present invention;

[0041] Figure 4This is a schematic diagram of the structure of a textbook data generation device provided in an embodiment of the present invention;

[0042] Figure 5 This is a schematic diagram of the structure of an electronic device that implements the textbook data generation method of this invention. Detailed Implementation

[0043] 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 should fall within the scope of protection of the present invention.

[0044] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0045] Figure 1 This is a flowchart of a textbook data generation method provided by an embodiment of the present invention. This embodiment is applicable to situations where textbooks are automatically generated and their content is automatically adjusted. This method can be executed by a textbook data generation device, which can be implemented in hardware and / or software and can be configured in an electronic device. Figure 1 As shown, the method includes:

[0046] S110. Generate initial teaching material data based on the lecture audio and video and the accompanying lesson plan data.

[0047] Specifically, initial teaching material data is generated based on the lecture audio / video and accompanying lesson plan data. First, Automatic Speech Recognition (ASR) tools are used to convert the lecture audio / video into timestamped text, ensuring precise alignment between the text content and the playback time of the audio / video and the time dimension of the accompanying lesson plan data (such as PowerPoint presentations). This guarantees the complete preservation of the original lecture information. Initial teaching material data is then automatically generated based on the timestamped text and the accompanying lesson plan data. For example, the teacher's lecture recordings, videos, and PowerPoint content are analyzed using technologies such as Natural Language Processing (NLP) and ASR to automatically organize and extract key points, generating a clearly structured and complete teaching material. This is similar to automatically organizing disorganized class notes into a well-organized electronic textbook, achieving automated transformation from unstructured lecture materials to structured teaching materials.

[0048] S120. Collect students' learning behavior data and homework data related to the audio and video lectures. Analyze the learning behavior data and homework data using a knowledge point assessment model built based on the Large Language Model (LLM) and output the assessment results of students' mastery of the knowledge points.

[0049] Specifically, learning behavior data includes at least one of the following: the distribution of time spent watching the lecture audio / video, the time points when notes were taken, the frequency and length of notes, and interaction data with the course content; homework data includes at least one of the following: homework completion time, homework accuracy rate, and difficulty level of the questions. The process of collecting students' learning behavior data on lecture audio / video and homework data and analyzing it through an LLM knowledge point assessment model is the core step in achieving personalized feedback. Learning behavior data is collected in real-time through classroom tools, including the distribution of time spent watching lecture audio / video and interaction with the course content; simultaneously, homework data is collected through question-and-answer tools, covering the completion time, accuracy rate, and difficulty level of assignments corresponding to each knowledge point. These two types of data are input into a knowledge point assessment model built on LLM. The model, through preset multi-dimensional assessment indicators and leveraging the semantic understanding and logical association capabilities of LLM, reflects students' actual mastery of different knowledge points.

[0050] S130. Input the mastery assessment results into the strategy optimization model built on LLM, output adjustment parameters of at least one dimension through the strategy optimization model, generate a textbook optimization strategy based on the adjustment parameters, and optimize the initial textbook data based on the textbook optimization strategy.

[0051] Specifically, after inputting the students' knowledge mastery assessment results into the strategy optimization model built on LLM, the model combines the assessment results with pre-set teaching patterns. Leveraging the semantic understanding and parameter tuning capabilities of LLM, it outputs adjustment parameters for at least one dimension, corresponding to different optimization dimensions of the textbook content. Based on these adjustment parameters, specific textbook optimization strategies are generated: for example, for knowledge points with weak mastery, the strategy explicitly increases their content weight; for knowledge points with strong mastery, the strategy decreases their weight. The initial textbook data is then structurally adjusted according to the optimization strategies, achieving a precise match between the textbook content and students' actual mastery, and driving the textbook's iteration from a static initial version to a dynamically adapted version.

[0052] Figure 2 This is a flowchart of an initial teaching material data generation method provided by an embodiment of the present invention. This embodiment further explains and illustrates the above embodiments. Figure 2 As shown, the method includes:

[0053] S210. Use a speech recognition tool to convert the lecture audio and video into lecture text with timestamps. The content of the lecture text corresponds to the specific playback time of the lecture audio and video.

[0054] Converting lecture audio and video into timestamped lecture text using speech recognition tools transforms unstructured lecture content into analyzable text. Specifically, speech recognition technology processes the speech signals in the lecture audio and video, converting the teacher's narration into written lecture text. Simultaneously, each generated text segment is automatically marked with a corresponding timestamp, indicating the specific playback time of that text segment in the original lecture audio and video, such as "00:05:30-00:06:15" corresponding to a segment about function definitions. This time correlation ensures a precise correspondence between the text content and the audio and video playback progress, fully preserving the core information of the lecture while establishing a correspondence between text segments and the original audio and video scene. This provides accurate temporal information for subsequent alignment with accompanying lesson plans (such as PowerPoint presentations), knowledge point extraction, and logical organization, ensuring the accurate reproduction of the classroom content in the subsequent textbook generation process.

[0055] S220. Input the teaching text and accompanying lesson plan data into the knowledge point assessment model, and output the extracted core teaching information of the classroom; determine the logical relationship between each knowledge point in the core teaching information; and generate a teaching syllabus framework based on the core teaching information and logical relationship.

[0056] Specifically, after inputting the timestamped lecture texts and accompanying lesson plans into the LLM-based analysis model, the model generates structured teaching outcomes through multi-level processing. First, the LLM, leveraging its semantic understanding capabilities, accurately extracts core classroom teaching information from the lecture texts, including teacher-spoken concept definitions, formula derivations, and case analyses, as well as from the accompanying lesson plans. This includes core subject concepts, key formulas and their derivation processes, theorem explanations, typical cases, and tips on common mistakes. It also needs to cover both the established key points in the lesson plans and the non-lesson plan content supplemented by the teacher in the lecture texts to ensure information completeness.

[0057] The model further analyzes the logical connections between various knowledge points in the core teaching information. By identifying causal words, progressive words, and inherent subject-specific rules in the text, it outlines the hierarchical, deductive, and cross-chapter connections between knowledge points, forming a visualized logical map of knowledge points. Based on the extracted core teaching information and the outlined logical relationships, the model generates a teaching syllabus framework in conjunction with the subject's teaching principles: clearly defining the chapter divisions of the textbook, the specific knowledge points contained in each chapter, and the weight of each knowledge point within the chapter. Furthermore, the chapter structure must be adapted to the course schedule to ensure that the syllabus reflects both the focus of classroom teaching and a clear logical progression of knowledge.

[0058] S230. Using the teaching syllabus framework as input, generate an agent by calling the content, plan the chapter structure of the textbook according to the teaching syllabus framework, and determine the length proportion of each chapter; based on the length proportion, transform the core teaching information into textbook text.

[0059] Specifically, after generating an agent by calling the content based on the teaching syllabus framework, the agent will complete the main body construction of the textbook according to the logic of structure, length, and text. First, following the chapter divisions clearly defined in the teaching syllabus framework, such as "Chapter 1 Basic Concepts, 1.1 Core Definitions, 1.2 Basic Properties," a clear hierarchical chapter structure that conforms to the teaching principles of the subject is planned, ensuring that the chapters are connected and logically connected without any gaps. Based on the weight of knowledge points in each chapter in the syllabus, the length allocation of the corresponding chapters is determined, with more space allocated to chapters containing core knowledge points to assist in the learning process. The chapters on key knowledge points are appropriately streamlined to ensure a precise match between the length of the text and the importance of the knowledge points. Finally, based on the determined length ratio, Agent transforms the previously extracted core classroom teaching information into textbook text. On the one hand, it transforms the colloquial expressions that may exist in the teaching text into rigorous language that conforms to the norms of the corresponding educational stage. On the other hand, it reasonably allocates the level of detail for each knowledge point according to the length ratio, emphasizing core information and keeping auxiliary information concise and clear. This ensures that the textbook text not only fully covers the core teaching content but also has a well-organized structure and appropriate level of detail, which is highly consistent with the requirements of the teaching syllabus framework.

[0060] S240. Determine the practice questions corresponding to the knowledge points of each chapter from the preset question bank, and obtain the initial textbook data based on the textbook text and practice questions.

[0061] Specifically, after constructing the textbook text, the content generation agent selects matching practice questions from a pre-set standardized question bank based on the core knowledge points of each chapter. The question bank is categorized and labeled according to knowledge point type, difficulty level, and question type. The agent accurately matches the appropriate question type based on the attributes of the chapter's knowledge points, initially prioritizing basic-level practice questions to ensure a strong correlation between the practice and the knowledge points, and that the difficulty level is suitable for students' initial learning level. Subsequently, the agent integrates and formats the generated textbook text with the selected practice questions according to a fixed structure of chapter titles, textbook text, and corresponding practice questions. The practice questions are clearly labeled with their corresponding knowledge points and question type categories, forming a complete and standardized initial textbook data.

[0062] In this embodiment of the invention, the question bank includes multiple question types, and each question type includes basic practice questions and advanced practice questions divided according to difficulty thresholds; correspondingly, practice questions corresponding to the knowledge points of each chapter are determined from the preset question bank, including:

[0063] Determine the suitability of each chapter's knowledge points with each type of question, and determine the first proportion of each type of question in each chapter's knowledge points based on the ratio of the suitability of each chapter's knowledge points with each type of question.

[0064] Based on the first proportion and the second proportion of pre-set basic and advanced practice questions, practice questions corresponding to the knowledge points of each chapter are determined from the pre-set question bank.

[0065] Specifically, the pre-set question bank has a clear classification structure, including multiple-choice questions, short-answer questions, calculation questions, case analysis questions, and other question types. Each question type is further divided into basic practice questions (suitable for the initial learning stage, with a higher difficulty level) and advanced practice questions (suitable for the consolidation and improvement stage, with a lower difficulty level) according to a preset difficulty threshold. Correspondingly, based on the attributes of the knowledge points in each chapter, the Agent calculates the fit between each chapter's knowledge points and each type of question through semantic matching and teaching suitability analysis. For example, conceptual knowledge points have a high fit with multiple-choice questions, and application-based knowledge points have a high fit with case analysis questions. Then, based on the ratio of the fit of different question types, the Agent determines the first proportion of each question type corresponding to the knowledge points in the total practice questions. For example, multiple-choice questions account for 60% and short-answer questions account for 40% in conceptual chapters. The Agent combines the preset fixed ratio of basic practice questions to advanced practice questions, i.e., the second proportion, such as 70% for initial basic practice questions and 30% for advanced practice questions. Using the first proportion as the basis for question type distribution and the second proportion as the basis for difficulty distribution, the Agent accurately selects practice questions from the question bank that are suitable in terms of quantity, question type, and difficulty. This ensures that the practice questions are highly consistent with the chapter's knowledge points and meet the difficulty suitability needs of students in the initial learning stage.

[0066] In this embodiment of the invention, after obtaining the teaching material text, the method further includes:

[0067] By using LLM to perform semantic validation on the textbook text, we can determine whether the textbook text conforms to the preset textbook specifications and whether the text content of each chapter matches the core teaching information.

[0068] Specifically, after generating the textbook text, LLM will perform a full scan of the textbook text according to preset textbook standards. These standards include: language style must conform to the requirements of the corresponding educational stage (primary and secondary school textbooks should be easy to understand, while higher education textbooks should be rigorous and professional); expression must avoid colloquialisms and ambiguities; and the definition of formulas and concepts must conform to subject standards. The LLM will then compare each chapter's textbook text with the previously extracted core teaching information point by point to verify whether the text fully covers the core information, whether the expression is consistent with the core information, and whether the logic matches the knowledge points in the core information. If any non-compliant or mismatched content is found, LLM will automatically mark it and provide feedback to the content generation agent, which will then make targeted corrections. Ultimately, this ensures that the textbook text not only meets the presentation requirements of standardized textbooks but also accurately reflects the core teaching information from the classroom.

[0069] Figure 3 This is a flowchart illustrating a method for outputting mastery assessment results according to an embodiment of the present invention. This embodiment further explains and clarifies the above embodiments. The learning behavior data includes the distribution of viewing time for lecture audio / video, the time points for note-taking, note frequency, and note length, as well as interaction data with the course content; the homework data includes homework completion time, homework accuracy rate, and the difficulty level of the questions. For example... Figure 3 As shown, the method includes:

[0070] S310. By using a knowledge point assessment model, the data on dwell time distribution, note-taking time, note frequency, note content length, and interaction behavior are weighted and evaluated to obtain a preliminary assessment result of mastery level.

[0071] Specifically, learning behavior data focuses on students' dynamic learning process with lecture audio and video. The distribution of time students spend watching lecture audio and video at different knowledge point segments reflects their level of attention to specific knowledge points; the specific time points when students take notes are related to the corresponding explanation content; the frequency of note-taking reflects learning initiative; and the length of the notes reflects the depth of understanding of knowledge points. There is also data on interaction behavior with the course content, such as highlighting key points, asking questions, and interactive discussions, reflecting participation and comprehension feedback. Homework data focuses on students' learning output, including the completion time of homework for each knowledge point, reflecting problem-solving proficiency; the homework accuracy rate, reflecting the effect of knowledge point mastery; and the difficulty level of the questions. Combining the accuracy rate with the homework data can further determine the solidity of mastery. For example, a low accuracy rate on basic questions indicates a weakness in core knowledge points.

[0072] The knowledge point assessment model assigns preset weights to dwell time distribution, note-taking time, note frequency, note content length, and interaction behavior data indicators. For example, dwell time distribution accounts for 40%, note-taking for 30%, and interaction behavior data for 30%. The weight allocation is determined based on the correlation between each indicator and the degree of knowledge point mastery. Leveraging the semantic understanding and data quantification capabilities of LLM, each dimension of data is scored concretely. For instance, in the dwell time distribution dimension, a high score is given if the dwell time for a particular knowledge point exceeds twice the average dwell time for the entire course, and a low score is given if it is less than 50%. In the note-taking dimension, a higher score is given if the note content length covers the core information of the knowledge point and the note-taking frequency matches the pace of the explanation. In the interaction behavior data dimension, a higher score is given for a high number of effective interactions such as highlighting key points and asking targeted questions. The model performs a weighted summation of scores for each dimension according to preset weights, outputting a quantitative score of 0-100. Based on the score, it categorizes students into three levels: below 60 points indicates weak initial mastery, 60-80 points indicates basic initial mastery, and above 80 points indicates proficiency in initial mastery. Ultimately, this forms a preliminary assessment result reflecting the student's initial mastery of each knowledge point.

[0073] S320. Based on the ratio of the completion time of assignments to the average completion time of assignments under different difficulty levels, and the ratio of the accuracy rate of assignments to the average accuracy rate of assignments, the preliminary mastery assessment results are verified and corrected to obtain the mastery assessment results.

[0074] Specifically, the knowledge point assessment model first processes homework data into strata based on the difficulty level of the questions, calculating two ratios: the ratio of a student's homework completion time at a certain difficulty level to the average completion time for questions at that level. A ratio < 1 indicates faster completion speed than average, reflecting problem-solving proficiency; a ratio > 1 indicates slower speed, potentially indicating a lack of solid understanding. The second ratio is the ratio of a student's homework accuracy rate at that difficulty level to the average accuracy rate for questions at that level. A ratio > 1 indicates higher-than-average accuracy, indicating good mastery; a ratio < 1 indicates insufficient accuracy, suggesting knowledge gaps. The model cross-validates these two stratified ratios with the initial mastery assessment results: if the initial assessment indicates proficiency, but the completion time ratio for high-difficulty questions is > 1 and the accuracy ratio is < 1, it indicates that actual mastery has not met expectations, and the assessment level will be lowered; if the initial assessment indicates basic mastery, but the completion time ratio for basic questions is < 1 and the accuracy ratio is > 1, and performance on advanced questions is stable, the assessment level will be raised; if both trends are consistent, the original level will be maintained and the scoring will be refined.

[0075] In this embodiment of the invention, an adjustment parameter of at least one dimension is output through a strategy optimization model, and a textbook optimization strategy is generated based on the adjustment parameter, including:

[0076] For knowledge points assessed as weak in the mastery assessment results, the content weighting parameters, explanation depth parameters, and length percentage parameters for that knowledge point will be adjusted, and the number of basic practice questions will be increased.

[0077] For knowledge points assessed as mastered or proficient in the mastery assessment results, the content weight adjustment parameters, explanation depth adjustment parameters, and length ratio adjustment parameters for that knowledge point will be reduced, and the number of advanced practice questions will be increased.

[0078] Specifically, the strategy optimization model is based on the semantic understanding and demand matching capabilities of LLM, and generates precise optimization strategies by using the assessment results of the mastery of knowledge points as a basis and through differentiated parameter configuration. The model first categorizes the assessment results, identifying the specific scope of knowledge points in two categories: weak mastery and proficient mastery. For each category, it outputs targeted adjustment parameters and translates them into practical strategies: For knowledge points assessed as weak mastery, the model outputs positive adjustment instructions to increase the content weight adjustment parameters, explanation depth adjustment parameters, and length proportion adjustment parameters. The corresponding optimization strategy is to increase the content proportion of this knowledge point in the textbook, such as increasing the explanation of core concepts, deepening the explanation level (e.g., adding formula derivation steps, adding typical example analyses, expanding the text length to enhance understanding), and increasing the number of basic practice questions from the question bank. For knowledge points assessed as proficient mastery, the model outputs negative adjustment instructions to decrease the content weight adjustment parameters, explanation depth adjustment parameters, and length proportion adjustment parameters. The corresponding optimization strategy is to streamline redundant explanations of this knowledge point, simplify repetitive explanation steps, reduce text length to avoid wasting information, and replace and increase the number of advanced practice questions from the question bank. By adjusting the parameters, the assessment results are directly transformed into a textbook optimization plan tailored to students' abilities, ensuring the targeted and effective optimization of the textbook.

[0079] Figure 4 This is a schematic diagram of the structure of a textbook data generation device provided in an embodiment of the present invention. Figure 4 As shown, the device includes:

[0080] The initial teaching material data generation unit 410 is used to generate initial teaching material data based on the lecture audio and video and the accompanying lesson plan data;

[0081] The evaluation result output unit 420 is used to collect students' learning behavior data and homework data on the teaching audio and video. The learning behavior data and homework data are analyzed by the knowledge point evaluation model built based on the large language model LLM, and the evaluation results of students' mastery of knowledge points are output.

[0082] The initial textbook optimization unit 430 is used to input the mastery assessment results into the strategy optimization model built on LLM, output adjustment parameters of at least one dimension through the strategy optimization model, generate textbook optimization strategies based on the adjustment parameters, and optimize the initial textbook data based on the textbook optimization strategies.

[0083] The textbook data generation device provided in this embodiment of the invention can execute the textbook data generation method provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the method.

[0084] Figure 5A schematic diagram of an electronic device 10 that can be used to implement embodiments of the present invention is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices (such as helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the invention described and / or claimed herein.

[0085] like Figure 5 As shown, the electronic device 10 includes at least one processor 11 and a memory, such as a read-only memory (ROM) 12 or a random access memory (RAM) 13, communicatively connected to the at least one processor 11. The memory stores computer programs executable by the at least one processor. The processor 11 can perform various appropriate actions and processes based on the computer program stored in the ROM 12 or loaded from storage unit 18 into the RAM 13. The RAM 13 can also store various programs and data required for the operation of the electronic device 10. The processor 11, ROM 12, and RAM 13 are interconnected via a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0086] Multiple components in electronic device 10 are connected to I / O interface 15, including: input unit 16, such as keyboard, mouse, etc.; output unit 17, such as various types of displays, speakers, etc.; storage unit 18, such as disk, optical disk, etc.; and communication unit 19, such as network card, modem, wireless transceiver, etc. Communication unit 19 allows electronic device 10 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0087] Processor 11 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, digital signal processors (DSPs), and any suitable processor, controller, microcontroller, etc. Processor 11 performs the various methods and processes described above, such as textbook data generation methods.

[0088] In some embodiments, the textbook data generation method may be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 18. In some embodiments, part or all of the computer program may be loaded and / or installed on electronic device 10 via ROM 12 and / or communication unit 19. When the computer program is loaded into RAM 13 and executed by processor 11, one or more steps of the textbook data generation method described above may be performed. Alternatively, in other embodiments, processor 11 may be configured to perform the textbook data generation method by any other suitable means (e.g., by means of firmware).

[0089] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), payload-programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0090] Computer programs used to implement the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when executed by the processor, the computer programs cause the functions / operations specified in the flowcharts and / or block diagrams to be performed. The computer programs may be executed entirely on a machine, partially on a machine, or as a standalone software package, partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0091] In the context of this invention, a computer-readable storage medium can be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, apparatus, or device. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination thereof. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0092] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0093] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or middleware components (e.g., application servers), or frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), blockchain networks, and the Internet.

[0094] A computing system can include clients and servers. Clients and servers are generally located far apart and typically interact through communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a hosting product within the cloud computing service system to address the shortcomings of traditional physical hosts and VPS services, such as high management difficulty and weak business scalability.

[0095] It should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps described in this invention can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this invention can be achieved, and this is not limited herein.

[0096] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. A method for generating textbook data, characterized in that, include: Initial teaching material data is generated based on the lecture audio and video and accompanying lesson plan data; The system collects students' learning behavior data and homework data related to the lecture audio and video recordings. It then analyzes this data using a knowledge point assessment model built on a Large Language Model (LLM) to output an assessment result of students' mastery of the knowledge points. The learning behavior data includes at least one of the following: the distribution of time spent watching the lecture audio and video recordings, the time points at which notes were taken, the frequency and length of notes, and interaction data with the course content. The homework data includes at least one of the following: homework completion time, homework accuracy rate, and the difficulty level of the questions. The mastery assessment results are input into a strategy optimization model based on LLM. The strategy optimization model outputs adjustment parameters for at least one dimension. A textbook optimization strategy is generated based on the adjustment parameters. The initial textbook data is then optimized based on the textbook optimization strategy.

2. The method according to claim 1, characterized in that, The initial teaching material data generated based on the lecture audio and video and accompanying lesson plan data includes: The speech recognition tool is used to convert the lecture audio and video into lecture text with timestamps, and the content of the lecture text corresponds to the specific playback time of the lecture audio and video; The teaching text and the accompanying lesson plan data are input into the knowledge point assessment model, and the extracted core classroom teaching information is output; the logical relationship between each knowledge point in the core teaching information is determined; and a teaching syllabus framework is generated based on the core teaching information and the logical relationship. Using the teaching syllabus framework as input, an intelligent agent is generated by calling the content. The chapter structure of the teaching materials is planned according to the teaching syllabus framework, and the length ratio of each chapter is determined. Based on the length ratio, the core teaching information is transformed into teaching material text. The practice questions corresponding to the knowledge points of each chapter are determined from the preset question bank, and the initial textbook data is obtained based on the textbook text and the practice questions.

3. The method according to claim 2, characterized in that, The question bank includes various question types, and each question type includes basic practice questions and advanced practice questions categorized according to difficulty thresholds; correspondingly, the step of determining the practice questions corresponding to each chapter's knowledge points from the preset question bank includes: Determine the fit between the knowledge points of each chapter and each question type, and determine the first proportion of each question type in each chapter based on the ratio of the fit between the knowledge points of each chapter and each question type; Based on the first proportion and the preset second proportion of basic and advanced practice questions, practice questions corresponding to the knowledge points of each chapter are determined from the preset question bank.

4. The method according to claim 2, characterized in that, After obtaining the textbook text, the method further includes: The textbook text is semantically validated using LLM to determine whether it conforms to the preset textbook specifications and whether the text content of each chapter matches the core teaching information.

5. The method according to claim 1, characterized in that, The learning behavior data and homework data are analyzed using a knowledge point assessment model built on LLM, and the results of the students' mastery of different knowledge points are output, including: The knowledge point assessment model is used to weight and evaluate the dwell time distribution, the time point of note-taking, the note frequency, the note content length, and the interaction behavior data to obtain a preliminary mastery assessment result. The preliminary mastery assessment results are verified and corrected based on the ratio of the completion time of the assignment to the average completion time of the assignment, and the ratio of the accuracy rate of the assignment to the average accuracy rate of the assignment, under different difficulty levels, to obtain the mastery assessment results.

6. The method according to claim 1, characterized in that, The step of outputting adjustment parameters of at least one dimension through the strategy optimization model, and generating a textbook optimization strategy based on the adjustment parameters, includes: For knowledge points assessed as weak in the mastery assessment results, the content weight adjustment parameters, explanation depth adjustment parameters, and length ratio adjustment parameters for that knowledge point are increased, and the number of basic practice questions is increased. For knowledge points assessed as mastered or proficient in the mastery assessment results, the content weight adjustment parameters, explanation depth adjustment parameters, and length ratio adjustment parameters for that knowledge point are reduced, while the number of advanced practice questions is increased.

7. A textbook data generation device, characterized in that, include: The initial teaching material data generation unit is used to generate initial teaching material data based on the lecture audio and video and the accompanying lesson plan data; The evaluation result output unit is used to collect students' learning behavior data and homework data related to the lecture audio and video. It analyzes the learning behavior data and homework data using a knowledge point evaluation model built based on the Large Language Model (LLM) to output an evaluation result of the students' mastery of the knowledge points. The learning behavior data includes the distribution of time spent watching the lecture audio and video, the time points at which notes were taken, the frequency of notes, the length of notes, and interaction data with the course content. The homework data includes homework completion time, homework accuracy rate, and the difficulty level of the questions. The initial textbook optimization unit is used to input the mastery assessment results into a strategy optimization model based on LLM, output adjustment parameters of at least one dimension through the strategy optimization model, generate a textbook optimization strategy based on the adjustment parameters, and optimize the initial textbook data based on the textbook optimization strategy.

8. An electronic device, characterized in that, The electronic device includes: At least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores a computer program executable by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the textbook data generation method according to any one of claims 1-6.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that are used to cause a processor to execute the textbook data generation method according to any one of claims 1-6.

10. A computer program product, characterized in that, The computer program product includes a computer program that, when executed by a processor, implements the textbook data generation method according to any one of claims 1-6.

Images