Intelligent English question setting method and system based on large model

Abstract

The invention provides an intelligent English question setting method and system based on a large model, and the method comprises the steps: obtaining all kinds of information in a teaching material, and generating a knowledge graph; extracting a dynamic knowledge graph; and constructing a question generation constraint, and generating test questions in combination with the dynamic knowledge graph. Through hierarchical and classified management and a dynamic screening mechanism of the knowledge graph, contents related to a current teaching target can be rapidly extracted, coverage and reasonable application of knowledge points are ensured, meanwhile, a new target can be rapidly adapted according to change of teaching materials, topic generation is restrained through the dynamic knowledge graph, and the teaching efficiency is improved. The generated questions are limited in the knowledge range of the students, the difficulty is dynamically adjusted according to the layering of knowledge points, the questions better conform to the teaching outline and the learning level of the students, teaching objectives can be effectively served, and meanwhile, the questions are subjected to multi-dimensional optimization by using a context enhancement and automatic evaluation mechanism, so that the teaching efficiency is improved. The method ensures that the questions are standard in language and clear in logic, and interference items of the selected questions are confused, so that the thinking investigation effect of students is improved.

Description

Technical Field

[0001] The present invention relates to the field of intelligent education technology, and in particular to an English intelligent question-setting method and system based on a large model. Background Art

[0002] Currently, the application of large language models (LLMs) in intelligent English question-setting for primary schools has demonstrated the ability to efficiently generate questions. It can generate various types of questions according to specified topics, contexts or difficulty requirements, such as vocabulary choice questions, grammar fill-in-the-blank questions, reading comprehension questions, etc., providing rich resources for teaching evaluation and practice, and greatly reducing the burden on teachers, but there are obvious problems.

[0003] English teaching usually has strict restrictions on students' vocabulary and knowledge points. However, the questions generated by the big model often contain difficult vocabulary or complex grammatical structures that are beyond the learning range of students, or topics that primary school students cannot understand. Coupled with the illusion problem of the model itself, many questions fail. In addition, the existing big model lacks the ability to accurately adapt to the course objectives, and the generated questions are less practical in teaching practice. Summary of the invention

[0004] The technical problem to be solved by the present invention is: in view of the shortcomings of the existing technology, a method and system for intelligent English question setting based on a big model is provided, which can divide the subject knowledge graph more finely, support dynamic updating, and integrate more closely with the big model.

[0005] In order to solve the above technical problems, the technical solution adopted by the present invention is: an English intelligent question-setting method based on a large model, comprising:

[0006] Obtain various information from English textbooks and generate knowledge graphs;

[0007] Extract dynamic knowledge graph;

[0008] Construct question generation constraints and generate test questions based on dynamic knowledge graphs.

[0009] Furthermore, the step of generating the knowledge graph includes:

[0010] Extract various types of information in English textbooks into triples to obtain information triple data;

[0011] Convert the vocabulary in the English textbook into different formats and save them to obtain vocabulary triple data;

[0012] Import information triple data and vocabulary triple data into the graph database to obtain the knowledge graph.

[0013] Furthermore, in the step of extracting a dynamic knowledge graph, it includes: obtaining question content, extracting relevant subgraphs from the knowledge graph according to the question content, and generating a dynamic knowledge graph.

[0014] Furthermore, in the step of constructing question generation constraints and generating test questions in combination with the dynamic knowledge graph, the following steps are included:

[0015] Obtain the question content and extract the prompt words in the question content;

[0016] Mark the knowledge points according to the prompt words to obtain the knowledge point information;

[0017] Substitute the knowledge point information into the preset question type template to obtain the question.

[0018] Furthermore, the step of obtaining the question content includes:

[0019] Use a large language model to perform semantic analysis on the question content to obtain query demand information;

[0020] Query the query requirement information according to the query model to obtain the query result;

[0021] Extract relevant subgraphs from the knowledge graph based on the query results to generate a dynamic knowledge graph.

[0022] Furthermore, before the step of querying the query demand information according to the query model to obtain the query result, the query model is also trained.

[0023] Furthermore, it also includes evaluating the test questions, obtaining evaluation results, and optimizing the generation strategy according to the evaluation results.

[0024] The present invention also relates to an English intelligent question-setting system based on a large model, comprising an acquisition module, an extraction module, a construction module and a generation module;

[0025] The acquisition module is used to acquire various types of information in the English teaching materials;

[0026] The extraction module is used to extract the dynamic knowledge graph;

[0027] The construction module is used to construct the question generation constraints;

[0028] The generation module is used to generate a knowledge graph and generate test questions.

[0029] The present invention also relates to an electronic device, comprising:

[0030] one or more processors;

[0031] A storage device for storing one or more programs;

[0032] When the one or more programs are executed by the one or more processors, the one or more processors implement any of the methods described above.

[0033] The present invention also relates to a computer-readable storage medium having a computer program stored thereon, wherein the program implements any of the above-mentioned methods when executed by a processor.

[0034] The beneficial effects of the present invention are: through the hierarchical classification management and dynamic screening mechanism of the knowledge graph, it is possible to quickly extract content related to the current teaching objectives from thousands of knowledge points, ensure the coverage and reasonable application of knowledge points, and at the same time, quickly change the knowledge graph according to changes in teaching materials to adapt to new goals. The dynamic knowledge graph constrains the generation of questions to ensure that the generated questions are strictly limited to the students' knowledge range, and dynamically adjust the difficulty according to the hierarchical level of knowledge points. The questions are more in line with the teaching syllabus and the students' learning level, and can effectively serve the teaching objectives. At the same time, the context enhancement and automated evaluation mechanism are used to optimize the questions in multiple dimensions to ensure that the question language is standardized, the logic is clear, and the interference items of the multiple-choice questions are confusing, thereby improving the effect of examining students' thinking. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] The above and other objects, features and advantages of the present disclosure will become more apparent by describing in detail its exemplary embodiments with reference to the accompanying drawings. The accompanying drawings described below are only some embodiments of the present disclosure, and it is clear to a person skilled in the art that other accompanying drawings can be obtained from these accompanying drawings without creative work.

[0036] Figure 1 It is a flow chart of the English intelligent question-setting method based on a large model of the present invention;

[0037] Figure 2 is a schematic diagram of the knowledge graph of the present invention;

[0038] Figure 3 A schematic diagram of the visual interface of the invention;

[0039] Figure 4 It is a schematic block diagram of the English intelligent question-setting system based on a large model of the present invention;

[0040] Figure 5 A schematic block diagram of an electronic device of the present invention. DETAILED DESCRIPTION

[0041] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in many forms and should not be construed as limited to the embodiments set forth herein; rather, these embodiments are provided so that the present invention will be comprehensive and complete and fully convey the concepts of the example embodiments to those skilled in the art. The same reference numerals in the figures represent the same or similar parts, and thus their repeated description will be omitted.

[0042] In addition, the described features, structures or characteristics may be combined in one or more embodiments in any suitable manner. In the following description, many specific details are provided to provide a full understanding of the embodiments of the present invention. However, those skilled in the art will appreciate that the technical solution of the present invention can be practiced without one or more of the specific details, or other methods, components, devices, steps, etc. may be adopted. In other cases, known methods, devices, implementations or operations are not shown or described in detail to avoid blurring various aspects of the present invention.

[0043] The block diagrams shown in the accompanying drawings are merely functional entities and do not necessarily correspond to physically independent entities. That is, these functional entities may be implemented in software form, or in one or more hardware modules or integrated circuits, or in different networks and / or processor devices and / or microcontroller devices.

[0044] The flowcharts shown in the accompanying drawings are only exemplary and do not necessarily include all the contents and operations / steps, nor must they be executed in the order described. For example, some operations / steps can be decomposed, and some operations / steps can be combined or partially combined, so the actual execution order may change according to actual conditions.

[0045] It should be understood that although the terms first, second, third, etc. may be used herein to describe various components, these components should not be limited by these terms. These terms are used to distinguish one component from another component. Therefore, the first component discussed below can be referred to as the second component without departing from the teachings of the present inventive concept. As used herein, the term "and / or" includes any one of the associated listed items and all combinations of one or more.

[0046] Those skilled in the art will appreciate that the drawings are merely schematic diagrams of example embodiments, and the modules or processes in the drawings are not necessarily necessary for implementing the present invention, and therefore cannot be used to limit the protection scope of the present invention.

[0047] Example

[0048] See also Figure 1 as well as Figure 2 , an English intelligent question-setting method based on a large model, comprising:

[0049] S1. Obtain various information from English textbooks and generate knowledge graphs;

[0050] By mapping various types of information in English textbooks into knowledge graphs and storing them in a graph database. The knowledge graph G = (V, E) is a directed graph, where:

[0051] G represents a directed graph, V is the set of nodes in the graph, and E is the set of edges.

[0052] The node set V contains various entity information and concepts in the textbook:

[0053] For example, grade information: indicates the grade corresponding to the textbook, such as first grade, second grade, third grade, etc.;

[0054] Textbook version information: such as People's Education Press, Shanghai Education Press, Foreign Language Teaching and Research Press, etc.;

[0055] Syllabus: includes the syllabus information of the textbook;

[0056] Unit: a specific unit or chapter in a textbook;

[0057] Vocabulary: key vocabulary in the textbook, such as verbs, nouns, adjectives, etc.;

[0058] Knowledge points: specific knowledge points or teaching objectives involved, such as grammar points such as morphology, syntax, and voice;

[0059] These nodes are distinguished and represented by appropriate identifiers (such as IDs, tags, etc.), and each node has its specific attribute information. For example, a vocabulary node contains information such as its meaning, usage, and example sentences;

[0060] The edge set E represents the relationship between entities. These relationships can be of various types, including:

[0061] The relationship between grades and textbook versions: that is, the textbooks corresponding to each grade under a certain version;

[0062] The textbook version is the root node, and its branches include all the textbook content, vocabulary and knowledge points in this system, which can easily distinguish different versions;

[0063] The hierarchical relationship between textbooks and units: it indicates the units included in a certain textbook version at a certain grade level;

[0064] The subordinate relationship between units and words: it indicates the words included in a unit. The first-grade textbook of the People's Education Press contains the words "book", "penci l", etc.

[0065] Dependency between vocabulary and knowledge points: For example, a certain vocabulary is associated with a certain grammar point, such as "run" is related to the knowledge point of "verb".

[0066] In addition to the above content, the knowledge graph also stores template question information, namely, multiple-choice questions, fill-in-the-blank questions, and reading question templates, which can provide reference for the large language model in the question-setting process.

[0067] S2, extract dynamic knowledge graph;

[0068] The dynamic knowledge graph G' is a subgraph related to the question content extracted from the global knowledge graph G. Specifically, the dynamic knowledge graph G' = (V', E') consists of the following two parts:

[0069] Node set V': contains the subset of entities such as grade, textbook, vocabulary, knowledge points, etc. related to the content of the question;

[0070] Edge set E': contains the subset of relations associated with the nodes in V'.

[0071] S3. Construct question generation constraints and generate test questions based on the dynamic knowledge graph.

[0072] Construct appropriate prompt words based on the given knowledge points and grade requirements, and generate test questions through the prompt words combined with the dynamic knowledge graph.

[0073] Q = Generate (G', P);

[0074] Where P is the question parameter, including multiple-choice questions, fill-in-the-blank questions, reading questions, etc.; Q is the set of generated questions, Q = {q1, q2, ..., qn}.

[0075] As a preferred embodiment of this embodiment, the step of generating a knowledge graph includes:

[0076] Extract various types of information in English textbooks into triples to obtain information triple data;

[0077] Convert the vocabulary in the English textbook into different formats and save them to obtain vocabulary triple data;

[0078] Import information triple data and vocabulary triple data into the graph database to obtain the knowledge graph.

[0079] In the specific implementation, in order to structure the storage and effectively represent the relationship between entities, the entities and their relationships in the knowledge graph are represented in the form of triples (h, r, t):

[0080] h (head entity): represents the starting node or main entity in the knowledge graph;

[0081] r (relation): indicates the semantic relationship between the head entity and the tail entity. For example, "contains", "belongs to", "depends on", "defines", etc.;

[0082] t (tail entity): represents another entity associated with the head entity;

[0083] These triples are stored in the knowledge graph, which enables more refined and hierarchical information management and query.

[0084] More specifically, the Graph RAG technology is used to extract the information contained in the textbooks and syllabus into triples and store them in an Excel table; the words are converted into triples using a Python script file, such as (Unit1, contains, Hello), and stored in an Excel table;

[0085] Then the triple information in the Excel table is imported into the Neo4j graph database in batches to form a complete knowledge graph.

[0086] As a preferred embodiment of this embodiment, the step of extracting a dynamic knowledge graph includes: obtaining question content, extracting relevant subgraphs from the knowledge graph according to the question content, and generating a dynamic knowledge graph.

[0087] In specific implementation, the dynamic knowledge graph G' is to extract the subgraph related to the question content from the global knowledge graph G: G' = (V', E'), where V' is a subset of grades, textbooks, vocabulary, and knowledge points related to the content of the questions, and E' is a subset of the relationships between these entities;

[0088] The construction process of G' is a screening function Filter(G, T, S) = G';

[0089] Among them, G is the complete knowledge graph, T is the knowledge point tested in the question, such as nouns, verbs, present tense, past tense, etc.; S is the knowledge level, such as third grade; only nodes that meet T and S and their related edges are selected.

[0090] As a preferred embodiment of this embodiment, in the step of constructing question generation constraints and generating test questions in combination with the dynamic knowledge graph, the following steps are included:

[0091] Obtain the question content and extract the prompt words in the question content;

[0092] Mark the knowledge points according to the prompt words to obtain the knowledge point information;

[0093] Substitute the knowledge point information into the preset question type template to obtain the question.

[0094] In specific implementation, appropriate prompt words are constructed according to given knowledge points and grade requirements, and the prompt words are used to guide the large model to generate questions.

[0095] The process of constructing the prompt word includes the following steps:

[0096] Knowledge point tagging: Construct corresponding prompt words based on the input knowledge point T and grade S. For example, if T is "verb tense" and S is grade three, the prompt words can be: "Write questions suitable for grade three based on 'simple present tense' and 'past tense'."

[0097] Question template generation: Generate questions based on preset question templates. That is, according to the required question type, obtain the template from the knowledge graph, add it to the prompt words, and provide reference for the large language model;

[0098] Context enhancement: Combine the dynamic knowledge graph G' with prompt words, and give certain examples according to the situation; for example, the textbook may contain instructions on the usage of "verbs", or have different contexts and themes. These contents can be integrated together to form a complete prompt word, so that the generated questions meet the teaching requirements.

[0099] Dynamically generate prompt words: Use students' answer history data to adjust prompt words to suit students' ability levels. For students with stronger abilities, prompt words can be more complex and involve more advanced grammatical structures; while for students with weaker abilities, prompt words will be simpler and more straightforward.

[0100] Distractor generation and optimization: Distractors are dynamically generated according to the context of the question, so that the options are semantically related to the correct answer but not completely correct, so as to achieve the purpose of testing students' mastery;

[0101] Finally, we get the question Q = Generate (G', P);

[0102] Where P is the question parameter, including multiple-choice questions, fill-in-the-blank questions, reading questions, etc.; Q is the set of generated questions, Q = {q1, q2, ..., qn}.

[0103] As a preferred embodiment of this invention, the step of obtaining the question content includes:

[0104] The question content is semantically analyzed through a large language model to obtain query requirement information.

[0105] In specific implementation, in order to more accurately extract relevant subgraphs from the knowledge graph G, a large language model and a graph convolutional network can be used to perform semantic analysis;

[0106] In the vast knowledge network, it will take a lot of time to manually search and select versions, grades, knowledge points and other limiting conditions. You can input the question content in text form. For example, enter "Please give five multiple-choice questions on the knowledge point of inversion sentences, at the level of third grade in elementary school."

[0107] Similarly, the STT speech-to-text model can also be used to support voice input of question content.

[0108] Input the question content into the large language model for semantic understanding and analysis to obtain query demand information that is convenient for knowledge graph query;

[0109] (T,S)=Analyse(Input,Prompt);

[0110] T is the knowledge point tested in the question: for example, inverted sentences; S is the knowledge level: for example, third grade of primary school; Input is natural language input; Prompt is the prompt word, which lets the big model know the task to be completed and extracts the required keywords.

[0111] Query the query requirement information according to the query model to obtain the query result;

[0112] Extract relevant subgraphs from the knowledge graph based on the query results to generate a dynamic knowledge graph.

[0113] As a preferred embodiment of this embodiment, before the step of querying the query demand information according to the query model to obtain the query result, it also includes training the query model.

[0114] In specific implementation, the query process of the knowledge graph is to train a graph convolutional network (GCN) to propagate information using the graph structure and capture the relationship between entities;

[0115] The core idea of ​​graph convolutional networks is to aggregate the information of node neighbors through convolution operations and pass it to the target node. The specific operation formula is as follows:

[0116]

[0117] in, represents the feature representation of node v at the kth layer; N(v) represents the set of neighbor nodes of node v; deg(v) represents the set of neighbor nodes of node v; W (k) is the weight matrix of the kth layer; σ is the activation function, b (k) is the bias term;

[0118] Query model training process: The query model is trained through supervised learning, with the goal of enabling the model to accurately extract subgraphs related to the question requirements from the knowledge graph G. During training, the input of the query model is the node and edge features of the knowledge graph G, and the output is the subgraph G' related to the given knowledge points and grade requirements; during the training process, the graph matching loss function is used to optimize the model so that it can filter out the correct nodes and edges from the complete graph:

[0119]

[0120] Among them, f(x i ,yi ) represents the node pair (x i ,y i ), y i is the actual target node, x i is the prediction node of the model;

[0121] The input natural language is simplified and refined through a large language model, and then passed into the trained graph convolutional network to obtain the subgraph G', completing the construction of the dynamic knowledge graph G'=Filter(G,T,S).

[0122] As a preferred embodiment of this embodiment, it also includes evaluating the test questions, obtaining evaluation results, and optimizing the generation strategy according to the evaluation results.

[0123] In the specific implementation, in order to ensure that the generated test questions are accurate and reasonable, teachers can evaluate and provide feedback to optimize the generated test questions. The evaluation function is:

[0124] Evaluate(Q,G',T)={Score_c, Score_d, Score_r};

[0125] Among them, Score_c is the knowledge point accuracy score, indicating whether the content of question Q is targeted at the required knowledge points; Score_d is the difficulty adaptation score, indicating whether the question is beyond the syllabus and whether it matches the current level of the students; Score_r is the quality score, indicating whether the question is reasonable and whether it achieves the purpose of testing students.

[0126] The evaluation results are input into the optimization model, the generation strategy is adjusted through reinforcement learning, and the question Q'=Optimize(Q, Evaluate(Q, G', T)) is regenerated according to the adjusted generation strategy.

[0127] Q' is the set of optimized questions, Q is the set of questions generated last time, G' is the dynamic knowledge graph, and T is the knowledge point tested by the questions.

[0128] See also Figure 3 In order to clearly display the knowledge graph to users, the entire knowledge graph can be displayed using front-end technology.

[0129] Node: Displays information such as grade, textbook version, knowledge points, etc., and uses different colors and shapes to distinguish different types of nodes;

[0130] Edge: Displays the relationship between different nodes, such as subordinate relationships, dependency relationships, etc. Users can click on a node to view all its related entities and relationships;

[0131] At the same time, it provides a convenient update module, which is implemented by using the add, delete, modify and query functions of the graph database, and refreshes the visual interface after completion.

[0132] See also Figure 3 , you can query, filter, ask questions and submit feedback through a visual interface.

[0133] a. Dynamic display: After selecting different textbook versions, grades and knowledge points, the interface will automatically update and display the corresponding knowledge graph.

[0134] b. Interactive function: Users can click on nodes in the knowledge graph to view detailed information and click on edges to understand the relationship between nodes.

[0135] c. Question generation button: In the "Query Filter and Question Generator" project, after the user enters the filter conditions, click the "Generate Question" button. The system generates questions based on the entered conditions and displays the generated question list on the interface for the user to preview. The system can also generate and export question files. At the same time, the previously generated question files can be uploaded to the visualization interface in the "File Upload Interface" project;

[0136] d. Quality Assessment Feedback Button: A "Feedback" button is provided on the question preview interface. If the quality does not meet the standard, you can select the reason for not meeting the standard, such as: vocabulary beyond the syllabus, unclear context, unreasonable options (multiple options or options are irrelevant to the question content), etc.;

[0137] Among them, c. Question generation module provides interface options:

[0138] Textbook version selection: Users can select different textbook versions from the drop-down menu, such as People's Education Press, Shanghai Education Press, etc.

[0139] Grade selection: Drop-down options from first grade to sixth grade.

[0140] Knowledge point selection: Select questions to generate knowledge points that need to be tested, such as grammar rules or common vocabulary.

[0141] Difficulty: Select Easy, Medium, or Hard to affect the difficulty of the generated questions.

[0142] Question type selection: Users can choose to generate multiple-choice questions, fill-in-the-blank questions, reading comprehension questions, or cloze questions.

[0143] Number of questions: Set the number of generated questions. The upper limit for multiple-choice and fill-in-the-blank questions is 10 questions, and the upper limit for reading comprehension and cloze questions is 1 article.

[0144] When there are many knowledge points, it will take time to search, so it supports natural language input and voice input. After the large model understands and analyzes, the drop-down box will automatically select the specified content; the brief layout and information of the visual interface are as follows Figure 3 As shown, it can greatly facilitate the user's operation.

[0145] From the above description, it can be seen that the beneficial effects of the present invention are: through the hierarchical classification management and dynamic screening mechanism of the knowledge graph, it is possible to quickly extract content related to the current teaching objectives from thousands of knowledge points, ensuring the coverage and reasonable application of knowledge points, and at the same time, it is possible to quickly change the knowledge graph according to changes in teaching materials to adapt to new goals. The dynamic knowledge graph constrains the generation of questions to ensure that the generated questions are strictly limited to the scope of students' knowledge, and dynamically adjust the difficulty according to the hierarchical knowledge points. The questions are more in line with the teaching syllabus and the students' learning level, and can effectively serve the teaching objectives. At the same time, the context enhancement and automated evaluation mechanism are used to optimize the questions in multiple dimensions to ensure that the question language is standardized, the logic is clear, and the interference items of the multiple-choice questions are confusing, thereby improving the effect of examining students' thinking.

[0146] See also Figure 4 As shown, the present invention also relates to an English intelligent question-setting system based on a large model, comprising an acquisition module, an extraction module, a construction module and a generation module;

[0147] The acquisition module is used to acquire various types of information in the English teaching materials;

[0148] The extraction module is used to extract the dynamic knowledge graph;

[0149] The construction module is used to construct the question generation constraints;

[0150] The generation module is used to generate a knowledge graph and generate test questions.

[0151] It should be noted that technical personnel in the relevant field can clearly understand that the specific implementation process of the above-mentioned teaching device based on the English word teaching platform and each unit can refer to the corresponding description in the aforementioned method embodiment. For the convenience and conciseness of the description, it will not be repeated here.

[0152] Figure 5 It is a block diagram of an electronic device according to an exemplary embodiment.

[0153] Refer to the following Figure 5 The electronic device according to this embodiment of the present invention will be described. Figure 5 The electronic device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present invention.

[0154] like Figure 5 As shown, the electronic device is in the form of a general computing device. The components of the electronic device may include, but are not limited to: at least one processing unit, at least one storage unit, a bus connecting different system components (including the storage unit and the processing unit), a display unit, etc.

[0155] The storage unit stores a program code, which can be executed by the processing unit, so that the processing unit executes the steps of various exemplary embodiments of the present invention described in the above-mentioned English intelligent question-setting method based on a large model. For example, the processing unit can execute the following steps: Figure 1 Follow the steps shown in .

[0156] The storage unit may include a readable medium in the form of a volatile storage unit, such as a random access memory unit (RAM) and / or a cache memory unit, and may further include a read-only memory unit (ROM).

[0157] The storage unit may also include a program / utility having a set (at least one) of program modules, such program modules including but not limited to: an operating system, one or more application programs, other program modules and program data, each of which or some combination may include an implementation of a network environment.

[0158] The bus may represent one or more of several types of bus structures, including a memory unit bus or memory unit controller, a peripheral bus, an accelerated graphics port, a processing unit, or a local bus using any of a variety of bus architectures.

[0159] The electronic device may also communicate with one or more external devices (e.g., keyboards, pointing devices, Bluetooth devices, etc.), may communicate with one or more devices that enable a user to interact with the electronic device, and / or may communicate with any device that enables the electronic device to communicate with one or more other computing devices (e.g., routers, modems, etc.). This communication may be performed through an input / output (I / O) interface. In addition, the electronic device may also communicate with one or more networks (e.g., local area networks (LANs), wide area networks (WANs), and / or public networks, such as the Internet) through a network adapter. The network adapter may communicate with other modules of the electronic device through a bus. It should be understood that, although not shown in the figure, other hardware and / or software modules may be used in conjunction with the electronic device, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems, etc.

[0160] Through the description of the above implementation, it is easy for those skilled in the art to understand that the example implementation described here can be implemented by software, or by software combined with necessary hardware. Therefore, the technical solution according to the implementation of the present invention can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.) or on a network, and includes several instructions to enable a computing device (which can be a personal computer, a server, or a network device, etc.) to perform the above method according to the implementation of the present invention.

[0161] A computer-readable medium, a program product for implementing the above-mentioned English intelligent question-setting method based on a large model according to an embodiment of the present invention, can adopt a portable compact disk read-only memory (CD-ROM) and include program code, and can be run on a terminal device, such as a personal computer. However, the program product of the present invention is not limited to this. In this document, a readable storage medium can be any tangible medium containing or storing a program, which can be used by or in combination with an instruction execution system, an apparatus or a device.

[0162] The program product may use any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. The readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples (non-exhaustive list) of readable storage media include: an electrical connection with one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0163] The computer readable storage medium may include a data signal propagated in a baseband or as part of a carrier wave, wherein a readable program code is carried. This propagated data signal may take a variety of forms, including but not limited to an electromagnetic signal, an optical signal, or any suitable combination of the above. The readable storage medium may also be any readable medium other than a readable storage medium, which may send, propagate, or transmit a program for use by an instruction execution system, an apparatus, or a device or used in combination with it. The program code contained on the readable storage medium may be transmitted with any appropriate medium, including but not limited to wireless, wired, optical cable, RF, etc., or any suitable combination of the above.

[0164] Program code for performing the operations of the present invention may be written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Java, C++, etc., and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user computing device, partially on the user device, as a separate software package, partially on the user computing device and partially on a remote computing device, or entirely on a remote computing device or server. In cases involving a remote computing device, the remote computing device may be connected to the user computing device through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computing device (e.g., through the Internet using an Internet service provider).

[0165] The computer-readable medium carries one or more programs. When the one or more programs are executed by a device, the computer-readable medium implements the following functions:

[0166] Obtain various information from English textbooks and generate knowledge graphs;

[0167] Extract dynamic knowledge graph;

[0168] Construct question generation constraints and generate test questions based on dynamic knowledge graphs.

[0169] Those skilled in the art will appreciate that the above modules may be distributed in the device according to the description of the embodiment, or may be changed accordingly and used in one or more devices that are different from the present embodiment. The modules and / or units and / or sub-units of the above embodiments may be combined into one module and / or unit and / or sub-unit, or may be further divided into multiple modules and / or units and / or sub-units and / or sub-modules.

[0170] Through the description of the above embodiments, it is easy for those skilled in the art to understand that the example embodiments described here can be implemented by software, or by software combined with necessary hardware. Therefore, the technical solution according to the embodiment of the present invention can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.) or on a network, and includes several instructions to enable a computing device (which can be a personal computer, a server, a mobile terminal, or a network device, etc.) to execute the method according to the embodiment of the present invention.

[0171] The exemplary embodiments of the present invention are specifically shown and described above. It should be understood that the present invention is not limited to the detailed structure, configuration or implementation method described herein; on the contrary, the present invention is intended to cover various modifications and equivalent configurations included in the spirit and scope of the appended claims.

[0172] In addition, the structures, proportions, sizes, etc. shown in this specification and the drawings are only used to match the contents disclosed in the specification for those skilled in the art to understand and read, and are not used to limit the limiting conditions for the implementation of the present invention, so they have no substantial technical significance. Any modification of the structure, change of the proportion relationship or adjustment of the size, without affecting the technical effects and purposes that can be achieved by the present invention, should still fall within the scope of the technical contents disclosed in the present invention. At the same time, the terms such as "on", "first", "second" and "one" quoted in this specification are only for the convenience of description, and are not used to limit the scope of the implementation of the present invention. The change or adjustment of their relative relationship should also be regarded as the scope of the implementation of the present invention without substantially changing the technical contents.

Claims

1. An English intelligent question-setting method based on a large model, comprising: Obtain various information from English textbooks and generate knowledge graphs; Extract dynamic knowledge graph; Construct question generation constraints and generate test questions based on dynamic knowledge graphs.

2. The method according to claim 1, characterized in that: The steps of generating a knowledge graph include: Extract various types of information in English textbooks into triples to obtain information triple data; Convert the vocabulary in the English textbook into different formats and save them to obtain vocabulary triple data; Import information triple data and vocabulary triple data into the graph database to obtain the knowledge graph.

3. The method according to claim 1, characterized in that: The step of extracting a dynamic knowledge graph includes: obtaining question content, extracting relevant subgraphs from the knowledge graph according to the question content, and generating a dynamic knowledge graph.

4. The method according to claim 1, characterized in that: The steps of constructing question generation constraints and generating test questions in combination with the dynamic knowledge graph include: Obtain the question content and extract the prompt words in the question content; Mark the knowledge points according to the prompt words to obtain the knowledge point information; Substitute the knowledge point information into the preset question type template to obtain the question.

5. The method according to claim 3 or 4, characterized in that: The steps of obtaining the question content include: Use a large language model to perform semantic analysis on the question content to obtain query demand information; Query the query requirement information according to the query model to obtain the query result; Extract relevant subgraphs from the knowledge graph based on the query results to generate a dynamic knowledge graph.

6. The method according to claim 5, characterized in that: Before the step of querying the query demand information according to the query model to obtain the query result, the query model is also trained.

7. The method according to claim 1, characterized in that: It also includes evaluating the test questions, obtaining evaluation results, and optimizing the generation strategy based on the evaluation results.

8. An English intelligent question-setting system based on a large model, characterized by: It includes an acquisition module, an extraction module, a construction module and a generation module; The acquisition module is used to acquire various types of information in the English teaching materials; The extraction module is used to extract the dynamic knowledge graph; The construction module is used to construct the question generation constraints; The generation module is used to generate a knowledge graph and generate test questions.

9. An electronic device, characterized in that: include: one or more processors; A storage device for storing one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the method according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.

Images