A teaching auxiliary design method and system based on BROKE framework prompt words

Abstract

The application discloses a teaching auxiliary design method and system based on a BROKE framework prompt word, relates to the technical field of teaching assistance, and comprises the following steps: obtaining a target teaching task corresponding to teaching auxiliary design, classifying the complexity of the target teaching task, and obtaining a simple teaching task or a complex teaching task; according to the complexity classification result of the target teaching task, configuring a corresponding BROKE framework element set, generating a structured filling template corresponding to the BROKE framework element set; receiving element information input by a user based on the structured filling template, generating a target BROKE framework prompt word suitable for the target teaching task according to the element information, and generating corresponding teaching auxiliary content based on the target BROKE framework prompt word. The application solves the problem of low accuracy caused by poor adaptability of the teaching auxiliary method based on the BROKE framework prompt word in the prior art.

Description

Technical Field

[0001] This invention relates to the field of instructional aids technology, and in particular to an instructional aid design method and system based on BROKE framework prompts. Background Technology

[0002] With the deep integration of artificial intelligence technology and educational scenarios, cue word engineering, as a core bridge connecting human needs and the output of AI models, has been widely applied in the field of instructional design, such as instructional resource generation, personalized learning guidance, and instructional evaluation assistance. The BROKE framework, as an emerging cue word design methodology, provides a structured definition of five elements—Background, Role, Objective, Key Results, and Evolve—ensuring the standardization and effectiveness of cue words, and is gradually becoming an important reference framework for cue word construction in instructional design.

[0003] In existing technologies, instructional design methods based on the BROKE framework have demonstrated initial application value. Their structured nature helps teachers clearly define teaching needs, making the teaching content generated by AI models more aligned with teaching objectives. However, in practical teaching scenarios, the core technical issues surrounding the simplification of the prompt word framework and its adaptation to teaching have not been effectively resolved, severely limiting the large-scale promotion and application effectiveness of this method.

[0004] On the one hand, the inherent complexity of the BROKE framework contradicts the efficiency requirements of teaching scenarios. Existing prompt word designs based on the BROKE framework require complete coverage of the detailed definitions of the five elements. For simple teaching tasks (such as explaining a single knowledge point or generating basic exercises), this constitutes significant redundancy, increasing teachers' preparation time and leading to excessively long prompt words and high model call token consumption. This is unfriendly to users with limited budgets, such as primary and secondary schools and educational institutions in remote areas. Furthermore, the framework requires users to possess basic professional knowledge of prompt word engineering, enabling them to accurately define the boundaries and expression standards of each element. However, most frontline teachers lack technical backgrounds and commonly exhibit issues such as missing elements and vague expressions, further raising the barrier to entry for the framework. Therefore, current teaching aids based on BROKE framework prompt words suffer from poor adaptability, resulting in low accuracy. Summary of the Invention

[0005] In view of this, the purpose of the present invention is to provide a teaching aid design method and system based on BROKE frame prompts, which aims to solve the problem of poor adaptability and low accuracy of existing teaching aid methods based on BROKE frame prompts.

[0006] One object of the present invention is to provide a teaching aid design method based on BROKE framework prompts, the method comprising:

[0007] Obtain the target teaching tasks corresponding to the instructional aid design, classify the target teaching tasks by complexity, and obtain simple teaching tasks or complex teaching tasks.

[0008] Based on the complexity classification results of the target teaching task, configure the corresponding BROKE framework element set and generate a structured filling template corresponding to the BROKE framework element set.

[0009] The system receives element information input by the user based on the structured filling template, generates target BROKE framework prompts that are adapted to the target teaching task based on the element information, and generates corresponding teaching aids based on the target BROKE framework prompts.

[0010] Furthermore, the above-mentioned instructional aid design method based on BROKE framework cue words further includes, after the step of generating target BROKE framework cue words adapted to the target teaching task based on element information:

[0011] The validity of the element information is verified, and redundant information is removed and language is simplified for the element information that passes the verification.

[0012] The steps of removing redundant information and simplifying language in the verified element information include:

[0013] Extract the core semantic content from the element information, filter out the decorative expressions that are irrelevant to the teaching objectives, adopt the abbreviation rules specific to the teaching scenario, simplify the core semantic content, and shorten the text length while retaining key information;

[0014] The estimated token consumption of prompt words formed by the combination of simplified element information is calculated. If the consumption exceeds a preset threshold, non-core modification information is further trimmed until the consumption is lower than the preset threshold.

[0015] Furthermore, in the above-mentioned instructional aid design method based on BROKE framework prompts, the step of classifying the target instructional task by complexity to obtain simple or complex instructional tasks includes:

[0016] Extract the core attribute features of the target teaching task. The core attribute features include the teaching objective level, the complexity of the task output form, the number of student participation dimensions, and the task completion time limit requirements.

[0017] The teaching objective level, the complexity of the task output form, the number of student participation dimensions, and the task completion time limit are quantified to obtain corresponding quantitative scores. The comprehensive complexity score of the target teaching task is calculated based on the weighted summation formula and the quantitative scores.

[0018] Obtain a historical teaching task classification dataset. Based on the comprehensive score distribution of simple and complex teaching tasks in the historical data, calculate the intersection point of the scores of the two types of tasks. Combine the grade level attributes of the current teaching scenario to correct the deviation of the midpoint and obtain the final classification threshold.

[0019] The overall complexity score of the target teaching task is compared with the final classification threshold, and the target teaching task is determined to be a simple teaching task or a complex teaching task based on the comparison results.

[0020] Furthermore, the above-mentioned instructional aid design method based on BROKE framework prompts includes the following steps: quantifying the hierarchical level of the instructional objective, the complexity of the task output form, the number of student participation dimensions, and the task completion time limit to obtain corresponding quantitative scores; and calculating the comprehensive complexity score of the target instructional task based on the weighted summation formula and the quantitative scores.

[0021] Based on the pre-defined hierarchical mapping rules for teaching objectives, the extracted teaching objective levels are transformed into a first quantitative score.

[0022] Construct a task output form complexity evaluation model, input task output form feature parameters, and output a second quantitative score;

[0023] The number of student participation dimensions required for the target teaching task is statistically analyzed, and a third quantitative score is allocated according to the number of participation dimensions.

[0024] The weighting coefficients are determined based on the task completion time requirements;

[0025] The weighted summation formula is: Overall Score = (First Quantitative Score × ... W 1+ Second quantitative score × W 2+ Third Quantitative Score × W 3) × weighting coefficient.

[0026] Furthermore, in the aforementioned instructional aid design method based on BROKE framework prompts, the step of using teaching scenario-specific abbreviation rules to simplify core semantic content and shorten text length while retaining key information includes:

[0027] Each has its own dedicated abbreviation dictionary and expression simplification rule set. The expression simplification rule set includes component omission rules, sentence reconstruction rules, and semantic merging rules. Different three-dimensional classification combinations correspond to different rule activation priorities.

[0028] Extract the three-dimensional scene parameters corresponding to the target teaching task. The three-dimensional scene parameters include learning stage parameters, subject parameters, and teaching target type parameters. Based on the three-dimensional scene parameters, match the corresponding target-specific abbreviation library and target expression simplification rule set, and determine the activation order of each rule in the target expression simplification rule set.

[0029] The core semantic content is segmented using a word segmentation tool, and core words that can match the full name in the target-specific abbreviation dictionary are selected, replaced with the corresponding abbreviations, and a preliminary simplified text is generated.

[0030] Based on the activation order of the target representation simplification rule set, representation simplification processing is performed on the initially simplified text.

[0031] Furthermore, in the above-mentioned instructional aid design method based on BROKE framework prompts, the construction of the dedicated abbreviation lexicon includes:

[0032] Collect teaching standards documents, core terminology in textbooks, and frequently used teaching expressions for all subjects and grade levels;

[0033] Core vocabulary is extracted using natural language processing technology. Combined with the information transmission efficiency requirements of teaching scenarios, a mapping relationship between full name, abbreviation, and applicable context is designed. After review and verification, a basic abbreviation lexicon is formed.

[0034] We build a proprietary abbreviation database by continuously collecting and dynamically updating frequently used and concise expressions from teaching practice.

[0035] Following the step of performing expression simplification processing on the initially simplified text based on the activation order of the target expression simplification rule set, the method further includes:

[0036] A teaching semantic integrity verification model is constructed. The simplified text is input into the model, and the model outputs an integrity score based on the preset core teaching information verification dimensions.

[0037] If the integrity score is lower than the preset threshold, the activation order and intensity of the expression simplification rules will be adjusted to reduce the scope of non-core components or retain some necessary modification components, and the simplification process will be re-executed.

[0038] Furthermore, in the above-mentioned instructional aid design method based on BROKE framework prompts, the step of adjusting the activation order and intensity of expression simplification rules if the completeness score is lower than a preset threshold includes:

[0039] Based on the detailed scoring of the core teaching information verification dimensions, we determine the semantic missing types and associated simplification rules, locate the missing dimensions, and trace the target simplification rules and specific function parameters that caused the missing information.

[0040] Based on the three-dimensional scene parameters, a rule adjustment priority matrix is ​​constructed, and the adjustment priority weights of the three types of simplified expression rules are assigned in combination with the characteristics of the learning stage, subject, and teaching objective type.

[0041] Based on the semantic missing type and associated expression simplification rules and the priority matrix dynamic reordering rules, the target simplification rules are shifted to the end, and non-target simplification rules are sorted by priority; if there are multiple target simplification rules, they are arranged in reverse order of semantic missing severity.

[0042] Establish a rule strength grading adjustment mechanism, classify missing levels according to the difference between the missing dimension score and the threshold, and match the corresponding rule strength adjustment coefficient.

[0043] Another object of the present invention is to provide a teaching aid design system based on BROKE framework prompts, the system comprising:

[0044] The data acquisition module is used to acquire the target teaching tasks corresponding to the instructional aid design, classify the target teaching tasks by complexity, and obtain simple or complex teaching tasks.

[0045] The configuration module is used to configure the corresponding BROKE framework element set according to the complexity classification result of the target teaching task, and generate a structured filling template corresponding to the BROKE framework element set.

[0046] The generation module is used to receive element information input by the user based on the structured filling template, generate target BROKE framework prompts adapted to the target teaching task based on the element information, and generate corresponding teaching aids based on the target BROKE framework prompts.

[0047] Another object of the present invention is to provide a readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method described above.

[0048] Another object of the present invention is to provide an electronic device including a memory, a processor, and a computer program stored in the memory and running on the processor, wherein the processor executes the program to implement the steps of the method described above.

[0049] This invention obtains the target teaching task corresponding to the instructional aid design, classifies the target teaching task by complexity to obtain simple or complex teaching tasks; based on the complexity classification result of the target teaching task, configures the corresponding BROKE framework element set, and generates a structured filling template corresponding to the BROKE framework element set; receives element information input by the user based on the structured filling template, generates target BROKE framework prompts adapted to the target teaching task based on the element information, and generates corresponding instructional aid content based on the target BROKE framework prompts. First, the target teaching task is classified by complexity to distinguish between simple and complex tasks. Differentiated configuration of BROKE framework elements is achieved through task complexity classification, avoiding redundant design under simple teaching tasks. Then, based on the classification result, a suitable BROKE framework element set is configured and a structured filling template is generated, guiding the user to input element information to generate prompts and instructional aid content adapted to the task. This solves the problem of poor adaptability and low accuracy in existing BROKE framework-based instructional aid methods. Attached Figure Description

[0050] Figure 1 This is a flowchart of a teaching aid design method based on BROKE framework prompts in the first embodiment of the present invention;

[0051] Figure 2 This is a structural block diagram of a teaching aid design system based on BROKE framework prompts according to the third embodiment of the present invention.

[0052] The following detailed description, in conjunction with the accompanying drawings, will further illustrate the present invention. Detailed Implementation

[0053] To facilitate understanding of the present invention, a more complete description will be given below with reference to the accompanying drawings. Several embodiments of the invention are illustrated in the drawings. However, the invention can be implemented in many different forms and is not limited to the embodiments described herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete.

[0054] It should be noted that when a component is said to be "fixed to" another component, it can be directly on the other component or there may be an intervening component. When a component is said to be "connected to" another component, it can be directly connected to the other component or there may be an intervening component. The terms "vertical," "horizontal," "left," "right," and similar expressions used in this document are for illustrative purposes only.

[0055] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. The terminology used herein in the description of the invention is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0056] Example 1

[0057] Please see Figure 1 The figure shows a teaching aid design method based on BROKE frame prompts in the first embodiment of the present invention, the method including steps S10 to S12.

[0058] Step S10: Obtain the target teaching task corresponding to the instructional aid design, classify the target teaching task by complexity, and obtain simple teaching task or complex teaching task.

[0059] The process involves identifying the target teaching tasks corresponding to the instructional aids design, categorizing these tasks by complexity to determine whether they are simple or complex. These target teaching tasks refer to specific teaching-related tasks that teachers need to complete using auxiliary tools during instructional preparation, such as "generating practice problems for two-digit multiplication in third-grade elementary school mathematics" or "designing an inquiry-based teaching plan for projectile motion in high school physics." The core purpose of complexity classification is to match different BROKE framework configurations based on the difficulty and complexity of the tasks, avoiding over-designing simple tasks that leads to inefficiency, while ensuring that the framework elements for complex tasks fully cover the needs. For example, explaining a single knowledge point or generating basic question types are simple teaching tasks; designing unit review plans or developing interdisciplinary project-based learning guidance plans are complex teaching tasks.

[0060] For example, to classify the complexity of target teaching tasks, core attribute features are extracted. These core attribute features include the level of teaching objectives, the complexity of task output form, the number of student participation dimensions, and the task completion time limit. The level of teaching objectives refers to the level of student ability development goals corresponding to the teaching task, such as different levels like understanding, comprehension, mastery, application, and innovation. The complexity of task output form refers to the complexity of the final output of the teaching task; for example, the complexity of a single practice question is lower than that of a combined output including practice questions, explanations, and extended thinking questions. The number of student participation dimensions refers to the dimensions students need to participate in when completing the learning activities related to the teaching task; for example, a task requiring only individual thinking has 1 participation dimension, while a task requiring group discussion, hands-on operation, and result presentation has 3. The task completion time limit refers to the time limit for completing the learning activities related to the teaching task; for example, a task completed within 10 minutes has a different time limit than a task completed within one class period.

[0061] Next, the levels of teaching objectives, the complexity of task output forms, the number of student participation dimensions, and the task completion time requirements are quantified to obtain corresponding quantitative scores. Based on a weighted summation formula and the quantitative scores, the comprehensive complexity score of the teaching task is calculated. Quantitative scoring transforms qualitative attributes into calculable values. For example, "understanding" at the teaching objective level is assigned 1 point, "comprehension" 2 points, "mastery" 3 points, "application" 4 points, and "innovation" 5 points. The complexity of task output forms can be assigned 1-5 points based on the type and number of output results and their production difficulty. The number of student participation dimensions is directly quantified according to the actual number, with 1 point for one dimension, 2 points for two dimensions, and so on. The weighted summation formula calculates the comprehensive complexity score by assigning different weights to different attribute characteristics. The weights are set based on the importance of each attribute to the complexity of the teaching task.

[0062] Next, a historical teaching task classification dataset was obtained. Based on the comprehensive score distribution of simple and complex teaching tasks in the historical data, the intersection point of the two types of task scores was calculated. The midpoint was then adjusted for deviation by considering the grade level attributes of the current teaching scenario, resulting in the final classification threshold. The historical teaching task classification dataset contains a large number of teaching tasks labeled as simple or complex and their corresponding comprehensive complexity scores. By analyzing the distribution of this data, the overlapping area of ​​simple and complex task scores was found, and the midpoint of the overlapping area was taken as the initial threshold. Different grade levels have different perceptions of teaching task complexity. For example, the score for a complex task in elementary school may be lower than the score for a simple task in middle school. Therefore, it is necessary to adjust the initial threshold based on grade level attributes (such as elementary, middle, and high school) to ensure that the classification threshold meets the actual needs of the current teaching scenario.

[0063] Finally, the overall complexity score of the target teaching task is compared with the final classification threshold. Based on the comparison results, the target teaching task is determined to be either a simple or complex teaching task. For example, if the final classification threshold is 3.5 points, and the overall complexity score of the target teaching task is 2.8 points, which is below the threshold, it is a simple teaching task; if the score is 4.2 points, which is above the threshold, it is a complex teaching task.

[0064] Furthermore, based on the pre-defined hierarchical mapping rules for teaching objectives, the extracted teaching objective levels are converted into a first quantitative score. The pre-defined hierarchical mapping rules for teaching objectives are correspondences established in advance according to the teaching syllabus and curriculum standards. For example, the mapping rules stipulate that "understanding" corresponds to 1 point, "comprehension" corresponds to 2 points, "mastery" corresponds to 3 points, "application" corresponds to 4 points, and "innovation" corresponds to 5 points. If the extracted teaching objective level is "mastery", then the first quantitative score is 3 points.

[0065] A task output form complexity evaluation model is constructed. Inputting task output form feature parameters, it outputs a second quantitative score. These parameters include the number of output types, production difficulty, and presentation format. For example, an output of "practice questions + explanations" has two types, a medium production difficulty, and a text presentation format. The evaluation model can be trained using machine learning algorithms, based on historical task output form feature parameters and manually labeled complexity scores. After training, inputting new feature parameters automatically outputs the corresponding second quantitative score; for example, inputting the above parameters results in a score of 3.

[0066] The number of student participation dimensions required for the target teaching task is statistically analyzed, and a third quantitative score is allocated based on the number of participation dimensions. The allocation rules can be preset as 1 point for 1 dimension, 2 points for 2 dimensions, and 3 points for 3 or more dimensions. For example, if the task requires students to engage in "group discussion + hands-on operation", the number of participation dimensions is 2, then the third quantitative score is 2 points.

[0067] The weighting coefficient is determined based on the task completion time limit. The rules for setting the weighting coefficient can be preset as follows: if the time limit is ≤10 minutes, the weighting coefficient is 0.8; if the time limit is 10 minutes < ≤30 minutes, the weighting coefficient is 1.0; if the time limit is >30 minutes, the weighting coefficient is 1.2. For example, if the task completion time limit is 20 minutes, the weighting coefficient is 1.0.

[0068] The weighted summation formula is: Overall Score = (First Quantitative Score × ...) W 1+ Second quantitative score × W 2+ Third Quantitative Score × W 3) × weight coefficients; where W1, W2, and W3 are the weights of the teaching objective level, the complexity of the task output form, and the number of student participation dimensions, respectively, which can be set according to the focus of the teaching scenario.

[0069] Step S11: Based on the complexity classification results of the target teaching task, configure the corresponding BROKE framework element set and generate a structured filling template corresponding to the BROKE framework element set.

[0070] Based on the complexity classification of the target teaching tasks, a corresponding BROKE framework element set is configured, generating a structured fill-in template corresponding to the BROKE framework element set. The BROKE framework element set is a combination of five elements: background, role, objective, key results, and optimization. Different element set configurations correspond to different task complexities—simple teaching tasks can simplify the detail of some elements, for example, retaining only the three core elements: role, objective, and key results; complex teaching tasks require a complete configuration of all five elements, and even further breakdown and refinement of some elements. The structured fill-in template transforms the configured element set into a standardized form that teachers can intuitively fill out, such as a table. The table columns are, in order, element name, filling instructions, example, and filling area, helping teachers without a technical background accurately understand the filling requirements of each element and lowering the barrier to entry.

[0071] In addition, in some optional embodiments of the present invention, after the step of generating the target BROKE framework prompt, a step of validating and optimizing the element information is added. The core purpose is to improve the accuracy and conciseness of the prompt, reduce the token consumption of model calls, and ensure that the core teaching information is not lost.

[0072] First, the validity of the element information is validated. For elements that pass validation, redundant information is removed and the language is simplified. Validation primarily checks whether the element information completely covers the configured BROKE framework elements, whether the information is clearly and unambiguously expressed, and whether it conforms to the basic norms of the current teaching scenario. For example, it checks whether key information such as grade, subject, and question type is filled in the "Generate Practice Questions" task. If missing, the user is prompted to supplement; if the expression is ambiguous, the user is guided to correct it. Redundant information removal and language simplification, while ensuring the accurate transmission of teaching needs, remove irrelevant expressions and reduce text length.

[0073] For example, specific steps for eliminating redundant information and simplifying language include extracting the core semantic content from the element information, filtering out decorative expressions irrelevant to the teaching objectives, and using teaching-specific abbreviation rules to simplify the core semantic content, thereby shortening the text length while retaining key information. The core semantic content refers to the key information that directly reflects the teaching needs; decorative expressions irrelevant to the teaching objectives can be directly filtered out. Teaching-specific abbreviation rules are simplification rules formulated for specific expressions in the teaching field, such as simplifying "third-grade elementary school math" to "elementary three math," and "basic practice questions" to "basic exercises," etc.

[0074] Subsequently, the estimated token consumption of the prompt words formed by the combination of simplified element information is calculated. If the consumption exceeds a preset threshold, non-core descriptive information is further pruned until the consumption falls below the preset threshold. A token is the basic unit for AI models to process text. Different models have different token consumption limits and billing standards. The preset threshold can be set according to the token limits of commonly used models and the user's budget requirements, for example, 200 tokens. The estimated token consumption can be achieved using existing text token calculation tools. If the estimated consumption of the simplified prompt words is 250 tokens, exceeding the 200-token threshold, non-core descriptive information is further pruned.

[0075] Step S12: Receive element information input by the user based on the structured filling template, generate target BROKE framework prompts adapted to the target teaching task based on the element information, and generate corresponding teaching aids based on the target BROKE framework prompts.

[0076] Specifically, the element information refers to the specific information related to the teaching task that the teacher fills in based on the template prompts. For example, for the task of "generating practice problems on two-digit multiplication for third-grade elementary school students," the teacher fills in the role as "elementary school math teacher" and the goal as "generating 10 basic practice problems and answers." The prompt word generation process transforms the structured element information into natural language text that conforms to the recognition standards of artificial intelligence models, ensuring that the model can accurately understand the teaching needs. The teaching support content is the output of the artificial intelligence model based on the prompt words, which can be expressed as practice problems, teaching plans, learning guidance scripts, evaluation scales, etc., depending on the task type.

[0077] In summary, the instructional aid design method based on BROKE framework prompts in the above embodiments of the present invention obtains the target teaching task corresponding to the instructional aid design, classifies the target teaching task by complexity to obtain simple or complex teaching tasks; configures the corresponding BROKE framework element set according to the complexity classification result of the target teaching task, and generates a structured filling template corresponding to the BROKE framework element set; receives element information input by the user based on the structured filling template, generates target BROKE framework prompts adapted to the target teaching task based on the element information, and generates corresponding instructional aid content based on the target BROKE framework prompts. First, the target teaching task is classified by complexity to distinguish between simple and complex tasks. Differentiated configuration of BROKE framework elements is achieved through task complexity classification, avoiding redundant design under simple teaching tasks. Then, based on the classification result, a suitable BROKE framework element set is configured and a structured filling template is generated, guiding the user to input element information to generate prompts and instructional aid content adapted to the task. This solves the problem of poor adaptability and low accuracy in existing BROKE framework-based instructional aid methods.

[0078] Example 2

[0079] This embodiment also proposes an instructional aid design method based on BROKE frame prompts. The difference between the instructional aid design method based on BROKE frame prompts in this embodiment and the instructional aid design method based on BROKE frame prompts in Embodiment 1 is as follows:

[0080] The steps of using teaching-specific abbreviation rules to simplify core semantic content and shorten text length while retaining key information include:

[0081] Each has its own dedicated abbreviation dictionary and expression simplification rule set. The expression simplification rule set includes component omission rules, sentence reconstruction rules, and semantic merging rules. Different three-dimensional classification combinations correspond to different rule activation priorities.

[0082] Extract the three-dimensional scene parameters corresponding to the target teaching task. The three-dimensional scene parameters include learning stage parameters, subject parameters, and teaching target type parameters. Based on the three-dimensional scene parameters, match the corresponding target-specific abbreviation library and target expression simplification rule set, and determine the activation order of each rule in the target expression simplification rule set.

[0083] The core semantic content is segmented using a word segmentation tool, and core words that can match the full name in the target-specific abbreviation dictionary are selected, replaced with the corresponding abbreviations, and a preliminary simplified text is generated.

[0084] Based on the activation order of the target representation simplification rule set, representation simplification processing is performed on the initially simplified text.

[0085] First, separate dedicated abbreviation libraries and expression simplification rule sets are established. The expression simplification rule set includes rules for omitting components, sentence structure reconstruction, and semantic merging, with different rule activation priorities corresponding to different three-dimensional classification combinations. Dedicated abbreviation libraries are established separately for different combinations of three-dimensional scene parameters. For example, the "Elementary School - Mathematics - Knowledge Consolidation" scene corresponds to one dedicated abbreviation library, while the "High School - Physics - Experimental Inquiry" scene corresponds to another. Component omission rules omit modifiers, adverbs, complements, etc., in sentences that do not affect the core semantics. For example, "Physics knowledge points suitable for second-year junior high school students" is omitted as "Second-year junior high school physics knowledge points." Sentence structure reconstruction rules transform complex sentences into simpler ones. For example, "I hope to generate materials to help students review" is reconstructed as "Generate student review materials." Semantic merging rules merge multiple semantically related expressions into a concise expression. For example, "Generate practice questions and generate analysis" is merged into "Generate practice questions and analysis." The rules for different three-dimensional classification combinations have different activation priorities. For example, in the "primary school-Chinese language-character teaching" scenario, the component omission rule is activated first; in the "high school-chemistry-equation derivation" scenario, the semantic merging rule is activated first.

[0086] Secondly, the three-dimensional scene parameters corresponding to the target teaching task are extracted. These parameters include grade level parameters, subject parameters, and teaching objective type parameters. Based on these parameters, a target-specific abbreviation library and a simplified rule set for target expression are matched, and the activation order of each rule in the simplified rule set is determined. Grade level parameters include primary school, junior high school, and senior high school; subject parameters include Chinese, mathematics, English, physics, and chemistry; and teaching objective type parameters include knowledge consolidation, ability enhancement, experimental inquiry, and project practice. For example, if the target teaching task is "to generate review exercises on quadratic equations in junior high school mathematics," the corresponding three-dimensional scene parameter is "junior high school - mathematics - knowledge consolidation." Based on this parameter, a target-specific abbreviation library and a simplified rule set for target expression are matched, and the rule activation priority in this scenario is determined as: component omission rules > semantic merging rules > sentence reconstruction rules.

[0087] Subsequently, a word segmentation tool is used to perform word segmentation on the core semantic content, screening out the core words that can match the full names in the target exclusive abbreviation library, replacing them with the corresponding abbreviated forms, and generating a preliminary streamlined text. The word segmentation tool can adopt existing mature natural language processing word segmentation tools, such as the jieba word segmentation tool, to split the core semantic content into single words or phrases. For example, "review exercises for quadratic equations of one variable in junior high school mathematics" is segmented into "junior high school / mathematics / quadratic equations of one variable / review / exercises", and then it is matched from the target exclusive abbreviation library that "junior high school" corresponds to "junior", and "review exercises" corresponds to "review questions". After replacement, the preliminary streamlined text "junior mathematics quadratic equations of one variable review questions" is generated.

[0088] Finally, based on the enabling order of the target expression simplification rule set, perform expression simplification processing on the preliminary streamlined text. For example, in the order of "constituent omission rule > semantic merging rule > sentence pattern reconstruction rule", first perform constituent omission on "junior mathematics quadratic equations of one variable review questions", and there are no redundant components; then perform semantic merging, and merge "junior mathematics" into "junior number"; finally, the streamlined text "junior number quadratic equations of one variable review questions" is obtained.

[0089] Exemplarily, the construction of the exclusive abbreviation library includes:

[0090] Collect teaching specification documents, core terms of textbooks and high-frequency teaching expressions for each academic stage and each subject;

[0091] Extract core words through natural language processing technology, design the mapping relationship of full name - abbreviated form - applicable context in combination with the requirements of teaching scenario information transmission efficiency, and form a basic abbreviation library after review and verification;

[0092] Dynamically update and construct the exclusive abbreviation library by continuously collecting high-frequency streamlined expressions in teaching practice;

[0093] After the step of performing expression simplification processing on the preliminary streamlined text based on the enabling order of the target expression simplification rule set, it further includes:

[0094] Construct a teaching semantic integrity verification model, input the text after simplification processing into the model, and the model outputs an integrity score based on the preset teaching core information verification dimension;

[0095] If the integrity score is lower than the preset threshold, adjust the enabling order and intensity of the expression simplification rules, reduce the elimination range of non-core components or retain some necessary modifying components, and re-perform the simplification processing.

[0096] The construction of the exclusive abbreviation database includes three sub-steps: First, collecting teaching standard documents, core terminology, and frequently used teaching expressions from textbooks for each subject and grade level; second, extracting core vocabulary through natural language processing technology, designing a mapping relationship between full name, abbreviation, and applicable context based on the information transmission efficiency requirements of teaching scenarios, and forming a basic abbreviation database after review and verification. For example, the core vocabulary "six elements of narrative writing" is extracted, and the mapping relationship is designed as "full name: six elements of narrative writing - abbreviation: six essentials of narrative writing - applicable context: primary school upper grades Chinese language teaching". After being reviewed and verified by experts in the field of teaching, it is included in the basic abbreviation database; third, the exclusive abbreviation database is dynamically updated by continuously collecting frequently used simplified expressions in teaching practice. For example, if it is found in teaching practice that "reading comprehension answering skills" is often simplified to "reading answering skills", then this mapping relationship is included in the exclusive abbreviation database for the corresponding scenario.

[0097] In addition, after performing expression simplification processing on the initially simplified text based on the activation order of the target expression simplification rule set, a teaching semantic integrity verification step is added: a teaching semantic integrity verification model is constructed, the simplified text is input into the model, and the model outputs an integrity score based on preset teaching core information verification dimensions. The teaching semantic integrity verification model is trained using machine learning algorithms, and the training data includes a large amount of simplified text and corresponding manually annotated integrity scores; the preset teaching core information verification dimensions include grade level information, subject information, teaching objective information, task type information, key parameter information (such as number of questions, duration), etc. The model scores by checking whether the simplified text completely covers these core information dimensions, with a score range of 0-100 points, where a higher score indicates greater semantic integrity.

[0098] If the completeness score is lower than the preset threshold, the activation order and intensity of the simplification rules are adjusted to reduce the scope of non-core component removal or retain some necessary modifying components, and the simplification process is repeated. The preset threshold can be set according to the accuracy requirements of the teaching scenario, for example, 80 points. If the completeness score of the simplified text "Review Questions on Quadratic Equations in Elementary Mathematics" is 75 points, which is lower than 80 points, the model analysis finds that the core information dimension of "number of questions" is missing. The reason is that the component omission rule has excessively removed relevant expressions. In this case, the rule intensity is adjusted to reduce the scope of component omission, retain the key parameter "10 questions", and it is simplified again to "10 Review Questions on Quadratic Equations in Elementary Mathematics". It is then input into the model for verification again until the score is higher than 80 points.

[0099] Specifically, if the integrity score is lower than a preset threshold, the steps for adjusting the activation order and intensity of the expression simplification rules include:

[0100] Based on the detailed scoring of the core teaching information verification dimensions, we determine the semantic missing types and associated simplification rules, locate the missing dimensions, and trace the target simplification rules and specific function parameters that caused the missing information.

[0101] Based on the three-dimensional scene parameters, a rule adjustment priority matrix is ​​constructed, and the adjustment priority weights of the three types of simplified expression rules are assigned in combination with the characteristics of the learning stage, subject, and teaching objective type.

[0102] Based on the semantic missing type and associated expression simplification rules and the priority matrix dynamic reordering rules, the target simplification rules are shifted to the end, and non-target simplification rules are sorted by priority; if there are multiple target simplification rules, they are arranged in reverse order of semantic missing severity.

[0103] Establish a rule strength grading adjustment mechanism, classify missing levels according to the difference between the missing dimension score and the threshold, and match the corresponding rule strength adjustment coefficient.

[0104] Specifically, based on the detailed scoring of the core teaching information verification dimensions, the semantic missing types and associated simplification rules are determined. Missing dimensions are located, and the target simplification rules and specific parameters that caused the missing dimensions are traced. The detailed scoring of the core teaching information verification dimensions refers to the model's specific score for each verification dimension. For example, "segment information 8 points, subject information 10 points, teaching objective information 8 points, task type information 10 points, key parameter information 5 points" (each with a maximum score of 10 points). The detailed scoring reveals that the semantic missing type is "key parameter information missing." Further tracing reveals that the component omission rule omits the key parameter "10 questions." The corresponding target simplification rule is the component omission rule, and the specific parameter is "omit all quantity-related modifiers."

[0105] Secondly, a rule adjustment priority matrix is ​​constructed based on 3D scene parameters, and adjustment priority weights are assigned to three types of simplified expression rules in combination with the characteristics of grade level, subject, and teaching objective type. The rule adjustment priority matrix uses 3D scene parameters as rows and columns, and the weight values ​​represent the adjustment priority of each rule in different scenarios. For example, in the "Primary School - Mathematics - Knowledge Consolidation" scenario, the adjustment priority weight of the component omission rule is 0.5, the semantic merging rule is 0.3, and the sentence reconstruction rule is 0.2; in the "High School - Physics - Experimental Inquiry" scenario, the adjustment priority weight of the semantic merging rule is 0.4, the component omission rule is 0.3, and the sentence reconstruction rule is 0.3. The higher the weight, the higher the adjustment priority.

[0106] Subsequently, based on the determined semantic missing type and the associated expression simplification rules and priority matrix dynamic rearrangement rules activation order, the target simplification rules are shifted to the end, and non-target simplification rules are sorted by priority; if there are multiple target simplification rules, they are arranged in reverse order of semantic missing severity. For example, in the "Primary School - Mathematics - Knowledge Consolidation" scenario, the target simplification rule is the component omission rule, and the adjusted rule activation order is semantic merging rule > sentence reconstruction rule > component omission rule; if there are two target simplification rules, the component omission rule and the semantic missing severity caused by the component omission rule is higher than that caused by the semantic merging rule, then the adjusted order is sentence reconstruction rule > semantic merging rule > component omission rule.

[0107] Finally, a rule strength grading adjustment mechanism is established, classifying missing levels based on the difference between the missing dimension score and the threshold, and matching corresponding rule strength adjustment coefficients. Missing levels are divided into three categories: minor, moderate, and severe. For example, a difference ≤ 2 indicates a minor missing level, with a corresponding adjustment coefficient of 0.8 (i.e., reducing the rule strength to 80% of its original level), such as reducing the proportion of omitted components, decreasing the scope of sentence reconstruction, or narrowing the semantic merging range; a difference of 2 < 4 indicates a moderate missing level, with a corresponding adjustment coefficient of 0.5; and a difference > 4 indicates a severe missing level, with a corresponding adjustment coefficient of 0.2.

[0108] In summary, the instructional aid design method based on BROKE framework prompts in the above embodiments of the present invention obtains the target teaching task corresponding to the instructional aid design, classifies the target teaching task by complexity to obtain simple or complex teaching tasks; configures the corresponding BROKE framework element set according to the complexity classification result of the target teaching task, and generates a structured filling template corresponding to the BROKE framework element set; receives element information input by the user based on the structured filling template, generates target BROKE framework prompts adapted to the target teaching task based on the element information, and generates corresponding instructional aid content based on the target BROKE framework prompts. First, the target teaching task is classified by complexity to distinguish between simple and complex tasks. Differentiated configuration of BROKE framework elements is achieved through task complexity classification, avoiding redundant design under simple teaching tasks. Then, based on the classification result, a suitable BROKE framework element set is configured and a structured filling template is generated, guiding the user to input element information to generate prompts and instructional aid content adapted to the task. This solves the problem of poor adaptability and low accuracy in existing BROKE framework-based instructional aid methods.

[0109] Example 3

[0110] Please see Figure 2The figure shows a teaching aid design system based on BROKE framework prompts proposed in the third embodiment of the present invention. The system includes:

[0111] The data acquisition module 100 is used to acquire the target teaching tasks corresponding to the instructional aid design, classify the target teaching tasks by complexity, and obtain simple teaching tasks or complex teaching tasks.

[0112] The configuration module 200 is used to configure the corresponding BROKE framework element set according to the complexity classification result of the target teaching task, and generate a structured filling template corresponding to the BROKE framework element set.

[0113] The generation module 300 is used to receive element information input by the user based on the structured filling template, generate target BROKE framework prompts adapted to the target teaching task based on the element information, and generate corresponding teaching aids based on the target BROKE framework prompts.

[0114] The functions or operation steps implemented by the above modules are largely the same as those in the above method embodiments, and will not be repeated here.

[0115] Example 4

[0116] In another aspect, the present invention provides a readable storage medium having a computer program stored thereon, wherein the program, when executed by a processor, implements the steps of the method described in any one of Embodiments 1 to 2 above.

[0117] Example 5

[0118] In another aspect, the present invention provides an electronic device, the electronic device including a memory, a processor, and a computer program stored in the memory and running on the processor, wherein the processor executes the program to implement the steps of any one of the methods described in Embodiments 1 to 2 above.

[0119] 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.

[0120] Those skilled in the art will understand that the logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequential list of executable instructions for implementing logical functions, and can be embodied in any computer-readable storage medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-included system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable storage medium" can mean any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device.

[0121] More specific examples (a non-exhaustive list) of computer-readable storage media include: electrical connections (electronic devices) having one or more wires, portable computer disk drives (magnetic devices), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Furthermore, computer-readable storage media can even be paper or other suitable media on which the program can be printed, since the program can be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in computer memory.

[0122] It should be understood that various parts of the present invention can be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0123] In the description of this specification, references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0124] 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 present 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 this patent should be determined by the appended claims.

Claims

1. A teaching aid design method based on BROKE framework prompts, characterized in that, The method includes: Obtain the target teaching tasks corresponding to the instructional aid design, classify the target teaching tasks by complexity, and obtain simple teaching tasks or complex teaching tasks. Based on the complexity classification results of the target teaching task, configure the corresponding BROKE framework element set and generate a structured filling template corresponding to the BROKE framework element set. Receive element information input by the user based on the structured filling template, generate target BROKE framework prompts adapted to the target teaching task based on the element information, and generate corresponding teaching aids based on the target BROKE framework prompts; Following the step of generating target BROKE framework prompts adapted to the target teaching task based on element information, the following is also included: The validity of the element information is verified, and redundant information is removed and language is simplified for the element information that passes the verification. The steps of removing redundant information and simplifying language in the verified element information include: Extracting the core semantic content from the element information, filtering out decorative expressions irrelevant to the teaching objectives, and employing abbreviation rules specific to teaching scenarios to streamline the core semantic content, thereby shortening the text length while retaining key information; specifically including: Each has its own dedicated abbreviation dictionary and expression simplification rule set. The expression simplification rule set includes component omission rules, sentence reconstruction rules, and semantic merging rules. Different three-dimensional classification combinations correspond to different rule activation priorities. Extract the three-dimensional scene parameters corresponding to the target teaching task. The three-dimensional scene parameters include learning stage parameters, subject parameters, and teaching target type parameters. Based on the three-dimensional scene parameters, match the corresponding target-specific abbreviation library and target expression simplification rule set, and determine the activation order of each rule in the target expression simplification rule set. The core semantic content is segmented using a word segmentation tool, and core words that can match the full name in the target-specific abbreviation dictionary are selected, replaced with the corresponding abbreviations, and a preliminary simplified text is generated. Based on the activation order of the target representation simplification rule set, representation simplification processing is performed on the initially simplified text; The estimated token consumption of prompt words formed by the combination of simplified element information is calculated. If the consumption exceeds a preset threshold, non-core modification information is further trimmed until the consumption is lower than the preset threshold.

2. The instructional aid design method based on BROKE framework prompts according to claim 1, characterized in that, The step of classifying the target teaching task by complexity to obtain simple or complex teaching tasks includes: Extract the core attribute features of the target teaching task. The core attribute features include the teaching objective level, the complexity of the task output form, the number of student participation dimensions, and the task completion time limit requirements. The teaching objective level, the complexity of the task output form, the number of student participation dimensions, and the task completion time limit are quantified to obtain corresponding quantitative scores. The comprehensive complexity score of the target teaching task is calculated based on the weighted summation formula and the quantitative scores. Obtain a historical teaching task classification dataset. Based on the comprehensive score distribution of simple and complex teaching tasks in the historical data, calculate the intersection point of the scores of the two types of tasks. Combine the grade level attributes of the current teaching scenario to correct the deviation of the midpoint and obtain the final classification threshold. The overall complexity score of the target teaching task is compared with the final classification threshold, and the target teaching task is determined to be a simple teaching task or a complex teaching task based on the comparison results.

3. The instructional aid design method based on BROKE framework prompts according to claim 2, characterized in that, The steps of quantifying the teaching objective level, task output complexity, number of student participation dimensions, and task completion time requirements to obtain corresponding quantitative scores, and calculating the comprehensive complexity score of the target teaching task based on the weighted summation formula and the quantitative scores, include: Based on the pre-defined hierarchical mapping rules for teaching objectives, the extracted teaching objective levels are transformed into a first quantitative score. Construct a task output form complexity evaluation model, input task output form feature parameters, and output a second quantitative score; The number of student participation dimensions required for the target teaching task is statistically analyzed, and a third quantitative score is allocated according to the number of participation dimensions. The weighting coefficients are determined based on the task completion time requirements; The weighted summation formula is: Overall Score = (First Quantitative Score × ... W 1+ Second quantitative score × W 2 + Third Quantitative Score × W 3) × weighting coefficient.

4. The instructional aid design method based on BROKE framework prompts according to claim 1, characterized in that, The construction of the proprietary abbreviation library includes: Collect teaching standards documents, core terminology in textbooks, and frequently used teaching expressions for all subjects and grade levels; Core vocabulary is extracted using natural language processing technology. Combined with the information transmission efficiency requirements of teaching scenarios, a mapping relationship between full name, abbreviation, and applicable context is designed. After review and verification, a basic abbreviation lexicon is formed. We build a proprietary abbreviation database by continuously collecting and dynamically updating frequently used and concise expressions from teaching practice. Following the step of performing expression simplification processing on the initially simplified text based on the activation order of the target expression simplification rule set, the method further includes: A teaching semantic integrity verification model is constructed. The simplified text is input into the model, and the model outputs an integrity score based on the preset core teaching information verification dimensions. If the integrity score is lower than the preset threshold, the activation order and intensity of the expression simplification rules will be adjusted to reduce the scope of non-core components or retain some necessary modification components, and the simplification process will be re-executed.

5. The instructional aid design method based on BROKE framework prompts according to claim 4, characterized in that, The step of adjusting the activation order and intensity of the expression simplification rules if the integrity score is lower than a preset threshold includes: Based on the detailed scoring of the core teaching information verification dimensions, we determine the semantic missing types and associated simplification rules, locate the missing dimensions, and trace the target simplification rules and specific function parameters that caused the missing information. Based on the three-dimensional scene parameters, a rule adjustment priority matrix is ​​constructed, and the adjustment priority weights of the three types of simplified expression rules are assigned in combination with the characteristics of the learning stage, subject, and teaching objective type. Based on the semantic missing type and associated expression simplification rules and the priority matrix dynamic reordering rules, the target simplification rules are shifted to the end, and non-target simplification rules are sorted by priority; if there are multiple target simplification rules, they are arranged in reverse order of semantic missing severity. Establish a rule strength grading adjustment mechanism, classify missing levels according to the difference between the missing dimension score and the threshold, and match the corresponding rule strength adjustment coefficient.

6. A teaching aid design system based on BROKE framework prompts, characterized in that, The system is used to implement the instructional aid design method based on BROKE framework prompts as described in any one of claims 1 to 5, the system comprising: The data acquisition module is used to acquire the target teaching tasks corresponding to the instructional aid design, classify the target teaching tasks by complexity, and obtain simple or complex teaching tasks. The configuration module is used to configure the corresponding BROKE framework element set according to the complexity classification result of the target teaching task, and generate a structured filling template corresponding to the BROKE framework element set. The generation module is used to receive element information input by the user based on the structured filling template, generate target BROKE framework prompts adapted to the target teaching task based on the element information, and generate corresponding teaching aids based on the target BROKE framework prompts.

7. A readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the steps of the method as described in any one of claims 1 to 5.

8. An electronic device, characterized in that, The method includes a memory, a processor, and a computer program stored in the memory and running on the processor, wherein the processor, when executing the program, implements the steps of the method as described in any one of claims 1 to 5.

Images