An online education training camp class test question pushing method, device and equipment

Abstract

The application discloses an online education training camp class test question pushing method, device and equipment, and relates to the technical field of artificial intelligence. The method is to automatically extract a current class new knowledge point set based on a current class text and a historical class text, then generate a plurality of class test questions corresponding to each new knowledge point in the set, and obtain the confidence of a target student in seriously listening to a class according to a class listening monitoring video segment in which a collection period is consistent with a display period of a corresponding knowledge point page of a lecture, and determine the corresponding and negatively related question quantity and / or question difficulty. Finally, at least one class test question is extracted according to the question quantity and / or the question difficulty and pushed to the target student. In this way, the class test question can be quickly and efficiently extracted, and the question paper can be ensured to have individuality for the student, thereby improving the question efficiency and quality.

Description

Technical Field

[0001] The present invention belongs to the field of artificial intelligence technology, and specifically relates to a method, device and equipment for pushing in-class test questions for an online education training camp. Background Art

[0002] As we all know, collective learning is far more effective than individual learning. The online education training camps launched by educational institutions embody a new hybrid online training model combining "online knowledge input with community services." By leveraging online communities to bring together students with similar learning needs, they not only enable efficient and systematic learning tasks but also create opportunities for communication and sharing, significantly enhancing the learning experience.

[0003] Currently, after completing a boot camp course, students are typically required to take an in-class test. These tests are primarily created by teachers or teaching teams, who then create the corresponding in-class exam papers. This process is both time-consuming and laborious. Furthermore, these exams are designed for all students and are not tailored to individual students.

[0004] With the continuous development of artificial intelligence technology, how to use AI (Artificial Intelligence) technology to automatically generate personalized in-class test questions in order to improve the efficiency and quality of question setting has become a topic that technical personnel in this field urgently need to study. Summary of the Invention

[0005] The purpose of the present invention is to provide a method, device, computer equipment, computer-readable storage medium and computer program product for pushing in-class test questions for an online education training camp, so as to solve the problems of low efficiency of manual question setting and lack of individual targeting for students in existing in-class test question setting solutions.

[0006] In order to achieve the above object, the present invention adopts the following technical solutions:

[0007] First, a method for pushing in-class test questions for an online education training camp is provided, comprising:

[0008] Obtain the current and historical lecture notes for the online education training camp courses;

[0009] Performing structural processing on the current class lecture text to obtain a current class lecture structured text, and also performing structural processing on the history class lecture text to obtain a history class lecture structured text;

[0010] Performing knowledge point extraction processing on the structured text of the current class lecture notes to obtain a current class knowledge point set, and further performing knowledge point extraction processing on the structured text of the history class lecture notes to obtain a history class knowledge point set;

[0011] remove the intersection of the current class knowledge point set and the historical class knowledge point set from the current class knowledge point set as a current class new learning knowledge point set;

[0012] For each new learning knowledge point in the current class new learning knowledge point set, a corresponding plurality of class test questions are generated;

[0013] For each new learning knowledge point, a class listening monitoring video segment with a collection period consistent with the display period of the corresponding knowledge point's lecture note page is intercepted from the target student's current class listening monitoring video, and the class listening monitoring video segment is imported into a listening state recognition model based on the first artificial intelligence algorithm and pre-trained, and the output is obtained. The first confidence of the target student's listening state on the corresponding knowledge point is in the serious listening state;

[0014] For each new learning knowledge point, according to the corresponding first confidence, the corresponding and negatively related question quantity and / or question difficulty are determined, and at least one class test question is extracted from the corresponding plurality of class test questions according to the question quantity and / or the question difficulty and pushed to the target student.

[0015] Based on the above invention content, a new scheme for automatically generating and pushing class test questions based on class lecture notes and class listening monitoring videos is provided, that is, based on the current class lecture note text and the historical class lecture note text of the online education training camp teaching course, the current class new learning knowledge point set is automatically extracted, then for each new learning knowledge point in the set, a plurality of corresponding class test questions are generated, and according to the class listening monitoring video segment with the collection period consistent with the display period of the corresponding knowledge point's lecture note page, the confidence of the target student's listening state on the corresponding knowledge point is obtained. In the serious listening state, and determine the corresponding and negatively related question quantity and / or question difficulty, and finally at least one class test question is extracted from the corresponding plurality of class test questions according to the question quantity and / or the question difficulty and pushed to the target student, so that not only the class test questions can be quickly and efficiently extracted, but also the test papers can be ensured to have individuality for students, improve the test efficiency and quality, and be convenient for practical application and popularization.

[0016] In one possible design, the current class lecture note structured text is processed for knowledge point extraction to obtain a current class knowledge point set, including:

[0017] Applying the extractive question-answering model LangChain to extract a first set of knowledge points from the structured text of the current lecture notes, and applying ChatGPT, a first language model based on the Transformer structure and that has been pre-trained, to generate a second set of knowledge points based on the structured text of the current lecture notes;

[0018] Matching to obtain the first knowledge point set and the knowledge point intersection of the first knowledge point set;

[0019] Applying RoBERTa, the second largest language model pre-built and pre-trained based on natural language inference technology, to determine whether there are any semantic incoherent parts in each knowledge point in the intersection of the knowledge points;

[0020] The intersection of the knowledge points from which the semantically incoherent parts have been removed is taken as the current in-class knowledge point set.

[0021] In one possible design, for each new knowledge point in the current set of new knowledge points to be learned in class, a plurality of corresponding in-class test questions are generated, including:

[0022] Using semantic matching technology to select a predefined question type template suitable for each new knowledge point in the set of new knowledge points to be learned in class, wherein the predefined question type template includes a question stem slot and an answer option slot, and the question types of the predefined question type template include multiple-choice questions, fill-in-the-blank questions, essay questions, true-or-false questions, and / or programming questions;

[0023] For each newly learned knowledge point, the corresponding knowledge point content is inserted into the question stem slot of the corresponding predefined question template to obtain the corresponding question stem content, and then a language generation model based on the GPT series language model is applied to generate answer options that meet the requirements of both grammar and semantics based on the question stem content, and the answer options are inserted into the answer option slot of the corresponding predefined question template to obtain the corresponding answer content;

[0024] For each newly learned knowledge point, assemble the corresponding question stem content and answer content to obtain the corresponding in-class test questions.

[0025] In one possible design, when there are multiple target students, for each newly learned knowledge point, at least one pop test question is extracted from the corresponding multiple pop test questions according to the corresponding question quantity and pushed to the target student, including:

[0026] For each of the newly learned knowledge points, traverse each of the target students in turn: according to the corresponding question quantity K determined for the currently traversed student, randomly select K unselected in-class test questions from the corresponding multiple in-class test questions, and push the extraction results to the currently traversed student, where K represents a positive integer.

[0027] In a second aspect, a device for pushing in-class test questions for an online education training camp is provided, comprising a lecture text acquisition unit, a structured processing unit, a knowledge point extraction unit, a newly learned knowledge point determination unit, a test question generation unit, a lecture status classification unit, and a test question extraction unit, which are sequentially connected in communication;

[0028] The lecture text acquisition unit is used to acquire the current class lecture text and the historical class lecture text of the online education training camp teaching course;

[0029] The structured processing unit is configured to perform structured processing on the current lecture text to obtain a structured text of the current lecture, and further perform structured processing on the history lecture text to obtain a structured text of the history lecture;

[0030] The knowledge point extraction unit is configured to perform knowledge point extraction processing on the current class lecture structured text to obtain a current class knowledge point set, and further perform knowledge point extraction processing on the history class lecture structured text to obtain a history class knowledge point set;

[0031] The newly learned knowledge point determination unit is configured to use the remaining set after removing the intersection of the current in-class knowledge point set and the historical in-class knowledge point set from the current in-class knowledge point set as the current in-class newly learned knowledge point set;

[0032] The test question generating unit is used to generate a plurality of corresponding in-class test questions for each new knowledge point in the current set of new knowledge points learned in class;

[0033] The listening state classification unit is configured to extract, from the target student's current in-class listening monitoring video, a segment of the in-class listening monitoring video whose acquisition period coincides with the display period of the handout page corresponding to the corresponding knowledge point, for each newly learned knowledge point, and import the segment of the in-class listening monitoring video into a pre-trained listening state recognition model based on a first artificial intelligence algorithm, and output a first confidence level that the target student's listening state at the corresponding knowledge point is an attentive listening state;

[0034] The test question extraction unit is used to determine, for each of the newly learned knowledge points, a corresponding number of questions and / or a question difficulty that is negatively correlated with the confidence level according to the corresponding first confidence level, and extract at least one in-class test question from the corresponding multiple in-class test questions based on the number of questions and / or the question difficulty, and push the result to the target student.

[0035] In a third aspect, the present invention provides a computer device comprising a memory, a processor, and a transceiver communicatively connected in sequence, wherein the memory is used to store a computer program, the transceiver is used to send and receive messages, and the processor is used to read the computer program and execute the method for pushing in-class test questions for an online education training camp as described in the first aspect or any possible design of the first aspect.

[0036] In a fourth aspect, the present invention provides a computer-readable storage medium having instructions stored thereon. When the instructions are executed on a computer, the method for pushing in-class test questions for an online education training camp as described in the first aspect or any possible design of the first aspect is executed.

[0037] In a fifth aspect, the present invention provides a computer program product, including a computer program or instructions, which, when executed by a computer, implements the method for pushing in-class test questions for an online education training camp as described in the first aspect or any possible design of the first aspect.

[0038] Beneficial effects of the above scheme:

[0039] (1) The present invention creatively provides a new solution for automatically generating and pushing in-class test questions based on in-class handouts and in-class lecture monitoring videos, that is, first, based on the current in-class handout text and the historical in-class handout text of the online education training camp teaching course, automatically extract the current in-class new knowledge point set, and then generate corresponding multiple in-class test questions for each new knowledge point in the set, and segment the in-class lecture monitoring video according to the acquisition period and the display period of the handout page to which the corresponding knowledge point belongs, to obtain the confidence that the target student is in a serious listening state at the corresponding knowledge point, and determine the corresponding number of questions and / or the difficulty of questions that are negatively correlated with the confidence, and finally extract at least one in-class test question from the corresponding multiple in-class test questions according to the number of questions and / or the difficulty of the questions and push it to the target student. In this way, not only can the in-class test questions be extracted quickly and efficiently, but also the test papers can be ensured to be targeted to the individual students, thereby improving the efficiency and quality of question setting and facilitating practical application and promotion. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] In order to more clearly illustrate the technical solutions of the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or prior art description. Obviously, the drawings in the following description only constitute some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained from these drawings without creative labor.

[0041] Figure 1 The flowchart of the online education training camp class test question pushing method provided by the embodiments of the present application.

[0042] Figure 2 The structural diagram of the online education training camp class test question pushing device provided by the embodiments of the present application.

[0043] Figure 3 The structural diagram of the computer device provided by the embodiments of the present application. DETAILED DESCRIPTION

[0044] In order to more clearly illustrate the technical solutions of the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or prior art description. Obviously, the drawings in the following description only constitute some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained from these drawings without creative labor. It should be noted that the description of these embodiment modes is used to help understand the present application, but does not constitute a limitation on the present application.

[0045] It should be understood that although the terms first and second, etc. may be used herein to describe various objects, these objects should not be limited by these terms. These terms are only used to distinguish one object from another. For example, the first object can be called the second object, and similarly, the second object can be called the first object, without departing from the scope of the example embodiments of the present application.

[0046] It should be understood that for the term "and / or" which may appear in the present application, it is only a description of the association relationship of the associated objects, which means that there can be three kinds of relationships, for example, A and / or B, which means that there are three kinds of situations, such as A alone, B alone, or A and B together; for example, A, B and / or C, which means that there are any one of A, B and C or any combination thereof; for the term " / and" which may appear in the present application, it is another description of the relationship of another associated object, which means that there can be two kinds of relationships, for example, A / and B, which means that there are two kinds of situations, such as A alone or A and B together; in addition, for the character " / " which may appear in the present application, it generally means that the associated objects before and after are an "or" relationship.

[0047] Example:

[0048] like Figure 1 As shown, the method for pushing test questions for online education training camps provided in the first aspect of this embodiment can be, but is not limited to, executed by a computer device with certain computing resources, such as a platform server, a personal computer (PC, a multi-purpose computer with a size, price and performance suitable for personal use; desktops, laptops, small laptops, tablets and ultrabooks are all personal computers), a smart phone, a personal digital assistant (PDA) or a wearable device. Figure 1 As shown, the method for pushing in-class test questions for an online education training camp may include, but is not limited to, the following steps S1 to S7.

[0049] S1. Obtain the current and historical lecture notes for the online education training camp courses.

[0050] In step S1, the online education training camp course is a teaching subject in an online education training camp teaching activity conducted for distributed students via the Internet, such as a programming subject. The current class handout text is the teaching material taught in the current online class, and the history class handout text is the teaching material taught in the previous and previous online classes. Both of them can be obtained through conventional upload by teaching staff.

[0051] S2. Structural processing is performed on the current class lecture text to obtain a current class lecture structured text, and also structural processing is performed on the history class lecture text to obtain a history class lecture structured text.

[0052] In step S2, the structured text is a line-based text file, which is a very common file format (such as TXT format or CSV format, etc.), and each line usually corresponds to a record, and each line has the same columns, which is equivalent to a data table in a database. In order to be able to complete the generation and push of the in-class test questions later, it is first necessary to convert the current / historical in-class handout text as a human language into a form that can be understood by the computer, so as to help the computer understand the meaning, structure and grammar of the text and extract useful information from it. Specifically, the current in-class handout text is structured to obtain the current in-class handout structured text, including but not limited to the following steps S21 to S23.

[0053] S21. Perform grammatical analysis on the current lecture text using natural language processing technology to obtain grammatical information including parts of speech and sentence structure.

[0054] In step S21, the natural language processing technology may include, but is not limited to, existing Tokenize technology (ie, word segmentation technology) and / or Part-of-speech tagging technology (ie, part-of-speech tagging technology).

[0055] S22. Identify at least one phrase representing a key knowledge point from the grammatical information using a rule / template matching method.

[0056] In step S22, the at least one phrase may be, but is not limited to, a noun phrase and / or a verb phrase.

[0057] S23. Using a predefined knowledge system template / ontology, determine the key knowledge points represented by each phrase in the at least one phrase, and obtain the current lecture structured text through structured representation.

[0058] In step S2, based on the aforementioned steps S21 to S23, the technical details of performing the structural processing on the history class lecture text to obtain the history class lecture structured text can also be derived, which will not be repeated here.

[0059] S3. Perform knowledge point extraction processing on the structured text of the current lecture notes to obtain a current lecture knowledge point set, and also perform knowledge point extraction processing on the structured text of the history lecture notes to obtain a history lecture knowledge point set.

[0060] In the step S3, the current in-class knowledge point set includes all the knowledge points explained in this online class. Specifically, in addition to the newly explained knowledge points, it may also include the knowledge points explained in the previous and previous online classes (i.e., the knowledge points explained repeatedly). The historical in-class knowledge point set includes all the knowledge points explained in the previous and previous online classes. In order to achieve the purpose of using AI (Artificial Intelligence) technology to complete knowledge point extraction and self-check whether there are self-consistency problems in knowledge points, preferably, the current in-class handout structured text is subjected to knowledge point extraction processing to obtain the current in-class knowledge point set, including but not limited to the following steps S31 to S34.

[0061] S31. Apply the extractive question-answering model LangChain to extract a first set of knowledge points from the structured text of the current lecture notes, and apply the first large language model ChatGPT based on the Transformer structure and that has been pre-trained to generate a second set of knowledge points based on the structured text of the current lecture notes.

[0062] In the step S31, the extraction question and answer model LangChain is a framework for developing applications based on language models, which can implement the following applications: data awareness, that is, connecting language models to other data sources; autonomy, that is, allowing language models to interact with their environment, so that the first set of knowledge points can be regularly extracted from the current class note structured text. The first large language model ChatGPT (full name: Chat Generative Pre-trained Transformer, a chat robot program developed by OpenAI, released on November 30, 2022, which is a natural language processing tool driven by artificial intelligence technology, which can generate answers based on patterns and statistical laws seen in the pre-training stage, and can interact according to the context of the chat, really like a human to chat, even can complete the tasks such as writing emails, video scripts, copywriting, translation, code or writing papers, etc.) can but not limited to specifically use GPT-4 model.

[0063] S32. Match the first set of knowledge points with the knowledge point intersection of the first set of knowledge points.

[0064] S33. Apply the second large language model RoBERTa pre-constructed based on natural language reasoning technology and pre-trained to determine whether there are semantic parts in each knowledge point in the knowledge point intersection.

[0065] In the step S33, the second large language model RoBERTa is an optimized version of the BERT (Bidirectional Encoder Representations from Transformers, a language representation model from Transformers) model, which has the advantage of using online dynamic masking technology in the pre-training process, rather than completing the masking operation in offline mode like BERT, so it does not require additional storage space and performs real-time masking replacement during model training, improving efficiency and accuracy. Thus, based on natural language reasoning technology and a certain amount of knowledge point positive and negative samples, the second large language model RoBERTa can be regularly trained to effectively identify whether there are semantic parts in the knowledge points.

[0066] S34. The knowledge point intersection with the knowledge points with semantic parts removed as the current class knowledge point set.

[0067] In step S3, based on the aforementioned steps S31 to S34, the technical details of obtaining a set of history class knowledge points by performing knowledge point extraction processing on the structured text of the history class lecture notes can also be derived, which will not be repeated here.

[0068] S4. The remaining set after removing the intersection of the current in-class knowledge point set and the historical in-class knowledge point set from the current in-class knowledge point set is used as the current in-class new knowledge point set.

[0069] In step S4, for example, if the current in-class knowledge point set includes knowledge point A, knowledge point B, knowledge point C, knowledge point D, knowledge point E and knowledge point F, and the historical in-class knowledge point set includes knowledge point A, knowledge point B, knowledge point D and knowledge point E, then the remaining set includes knowledge point C and knowledge point F, which are all new knowledge points learned in the current class.

[0070] S5. Generate a plurality of corresponding in-class test questions for each new knowledge point in the current set of new knowledge points learned in class.

[0071] In step S5, in order to achieve the purpose of generating the multiple in-class test questions of different question types using AI technology, preferably, for each new knowledge point in the current set of new knowledge points learned in class, corresponding multiple in-class test questions are generated, including but not limited to the following steps S51 to S53.

[0072] S51. Apply semantic matching technology to select predefined question type templates suitable for each new knowledge point in the current set of new knowledge points to be learned in class, wherein the predefined question type templates include but are not limited to question stem slots and answer option slots, etc., and the question types of the predefined question type templates cover but are not limited to multiple-choice questions, fill-in-the-blank questions, essay questions, true-or-false questions and / or programming questions, etc.

[0073] S52. For each newly learned knowledge point, the corresponding knowledge point content is inserted into the question stem slot of the corresponding predefined question type template to obtain the corresponding question stem content, and then a language generation model based on the GPT series language model is applied to generate answer options that meet the requirements in both grammar and semantics according to the question stem content, and the answer options are inserted into the answer option slot of the corresponding predefined question type template to obtain the corresponding answer content.

[0074] In step S52, the GPT series language model may be, but is not limited to, a GPT-4 model.

[0075] S53. For each newly learned knowledge point, assemble the corresponding question stem content and answer content to obtain the corresponding in-class test questions.

[0076] After step S53, in order to verify the usability of the in-class test questions, that is, to verify whether the in-class test questions contain conceptual errors (for example, the question itself is wrong), factual errors (for example, the correct answer options are already reflected in the question stem), logical errors (for example, there are no correct answer options) and / or grammatical errors (for example, the question description is incomplete or unclear), etc., it is further preferred that after obtaining the in-class test questions, the method further includes but is not limited to the following steps S541 to S543.

[0077] S541. Apply a test question error recognition model that is based on a second artificial intelligence algorithm and has completed pre-training to provide a second confidence level for whether the in-class test questions contain errors, wherein the test question error recognition model is used to identify whether the test questions contain errors, and the errors include but are not limited to conceptual errors, factual errors, logical errors and / or grammatical errors, etc.

[0078] In the step S541, the artificial intelligence algorithm is a core artificial intelligence algorithm that specifically studies how computers simulate or implement human learning behavior to acquire new knowledge or skills, reorganize existing knowledge structures and continuously improve their own performance. It is the fundamental way to make computers intelligent; specifically, the second artificial intelligence algorithm can be implemented by, but is not limited to, machine learning algorithms such as support vector machines, K-nearest neighbor method, stochastic gradient descent method, multi-layer perceptron, decision tree, back propagation neural network or radial basis function network, and can be pre-trained to obtain the test question error recognition model through conventional calibration verification modeling method (the specific process includes the calibration process and verification process of the model, that is, first comparing the model simulation results with the measured data, and then adjusting the model parameters according to the comparison results to make the simulation results consistent with the actual process).

[0079] S542. Determine whether the second confidence level exceeds a preset confidence level threshold.

[0080] In step S542 , the confidence threshold may be 61.8% for example.

[0081] S543. If yes, remove the in-class test question.

[0082] S6. For each newly learned knowledge point, extract a segment of the target student's current in-class monitoring video, the acquisition period of which is consistent with the display period of the handout page corresponding to the knowledge point, and import the segment into a pre-trained listening state recognition model based on a first artificial intelligence algorithm, and output a first confidence level that the target student's listening state at the corresponding knowledge point is a serious listening state.

[0083] In step S6, the target student is the subject to be assessed in the online class, and the current in-class listening monitoring video is a conventionally recorded monitoring video of the target student's listening performance in the online class (which can be conventionally recorded by the student's camera and uploaded locally). For example, if the presentation period of the handout page of a certain new knowledge point in the online class is from 10:00 to 10:05, then for the certain new knowledge point, it is necessary to extract the in-class listening monitoring video segment between 10:00 and 10:05 from the current in-class listening monitoring video of the target student. Since the in-class listening monitoring video segment will present the micro-expression characteristics of the target student during the presentation period of the handout page: behavioral characteristics such as concentration or distraction, the listening state recognition model can be pre-trained based on a certain amount of in-class listening monitoring video segment samples and sample labels (for example, attentive listening state or non-attentive listening state) through conventional calibration and verification modeling methods. In addition, specifically, the first artificial intelligence algorithm can be, but is not limited to, a machine learning algorithm based on a support vector machine, a decision tree, or a random forest.

[0084] S7. For each of the newly learned knowledge points, determine the corresponding number of questions and / or the difficulty of the questions that are negatively correlated with the confidence level based on the corresponding first confidence level, and extract at least one in-class test question from the corresponding multiple in-class test questions based on the number of questions and / or the difficulty of the questions and push it to the target student.

[0085] In step S7, the purpose of establishing the negative correlation between the first confidence level and the number of questions and / or the difficulty of the questions is: if the target student listens to the class more attentively on a certain new knowledge point (that is, the higher the first confidence level), it can be reflected that the target student has a higher level of mastery of the certain new knowledge point, and therefore fewer test questions or easier test questions can be pushed for routine in-class testing; and if the target student listens to the class less attentively on a certain new knowledge point (that is, the lower the first confidence level), it can be reflected that the target student has a lower level of mastery of the certain new knowledge point, and therefore more test questions or more difficult test questions can be pushed for focused in-class testing, so as to improve the target student's mastery of the certain new knowledge point, and then through this targeted question-setting method, the mastery of the new knowledge points of all students can be balanced, which is beneficial for them to keep up with the subsequent online education training camp teaching courses and improve the activity effect of the online education training camp. In addition, considering that the trainees in the training camp are not gathered together for teaching and training, there is a problem of difficulty in test supervision. Therefore, in order to avoid the phenomenon of multiple students copying answers from each other with the help of Internet communication technology (for example, through WeChat), preferably, when there are multiple target students, for each newly learned knowledge point, at least one in-class test question is extracted from the corresponding multiple in-class test questions according to the corresponding question quantity and pushed to the target student, including but not limited to: for each newly learned knowledge point, traverse each of the target students in turn: according to the corresponding and determined question quantity K for the current traversal student, randomly extract K unextracted in-class test questions from the corresponding multiple in-class test questions, and push the extraction results to the current traversal student, wherein K represents a positive integer. In this way, it can be ensured that the in-class test questions on the same newly learned knowledge point are different for each student, thereby effectively preventing them from copying answers from each other.

[0086] Therefore, based on the method for pushing in-class test questions for an online education training camp described in the aforementioned steps S1 to S7, a new solution for automatically generating and pushing in-class test questions based on in-class handouts and in-class lecture monitoring videos is provided. That is, first, based on the current in-class handout text and the historical in-class handout text of the online education training camp teaching course, a set of new knowledge points learned in the current class is automatically extracted. Then, for each new knowledge point in the set, multiple corresponding in-class test questions are generated. In addition, based on the in-class lecture monitoring video segmentation that coincides with the display period of the handout page to which the corresponding knowledge point belongs, a confidence level that the target student's listening status at the corresponding knowledge point is a diligent listening state is obtained, and a corresponding number of questions and / or question difficulty that are negatively correlated with the confidence level is determined. Finally, based on the number of questions and / or the question difficulty, at least one in-class test question is extracted from the corresponding multiple in-class test questions and pushed to the target student. In this way, not only can the in-class test questions be extracted quickly and efficiently, but also the test papers can be ensured to be individually targeted to the students, thereby improving the efficiency and quality of question setting and facilitating practical application and promotion.

[0087] like Figure 2 As shown, the second aspect of this embodiment provides a virtual device for implementing the method for pushing in-class test questions for an online education training camp described in the first aspect, comprising a lecture text acquisition unit, a structured processing unit, a knowledge point extraction unit, a newly learned knowledge point determination unit, a test question generation unit, a lecture status classification unit, and a test question extraction unit that are sequentially communicatively connected;

[0088] The lecture text acquisition unit is used to acquire the current class lecture text and the historical class lecture text of the online education training camp teaching course;

[0089] The structured processing unit is configured to perform structured processing on the current lecture text to obtain a structured text of the current lecture, and further perform structured processing on the history lecture text to obtain a structured text of the history lecture;

[0090] The knowledge point extraction unit is configured to perform knowledge point extraction processing on the current class lecture structured text to obtain a current class knowledge point set, and further perform knowledge point extraction processing on the history class lecture structured text to obtain a history class knowledge point set;

[0091] The newly learned knowledge point determination unit is configured to use the remaining set after removing the intersection of the current in-class knowledge point set and the historical in-class knowledge point set from the current in-class knowledge point set as the current in-class newly learned knowledge point set;

[0092] The test question generating unit is used to generate a plurality of corresponding in-class test questions for each new knowledge point in the current set of new knowledge points learned in class;

[0093] The listening state classification unit is configured to extract, from the target student's current in-class listening surveillance video, a segment of the in-class listening surveillance video whose acquisition period coincides with the display period of the handout page corresponding to the corresponding knowledge point, for each newly learned knowledge point, and import the segment of the in-class listening surveillance video into a pre-trained listening state recognition model based on a first artificial intelligence algorithm, and output a first confidence level that the target student's listening state at the corresponding knowledge point is an attentive listening state;

[0094] The test question extraction unit is used to determine, for each of the newly learned knowledge points, a corresponding number of questions and / or a question difficulty that is negatively correlated with the confidence level according to the corresponding first confidence level, and extract at least one in-class test question from the corresponding multiple in-class test questions based on the number of questions and / or the question difficulty, and push the result to the target student.

[0095] The working process, working details and technical effects of the aforementioned device provided in the second aspect of this embodiment can be referred to the method for pushing in-class test questions for the online education training camp described in the first aspect, and will not be described in detail here.

[0096] like Figure 3 As shown, the third aspect of the present embodiment provides a computer device for executing the online education training camp in-class test question pushing method as described in the first aspect, including a memory, a processor and a transceiver that are sequentially connected in communication, wherein the memory is used to store a computer program, the transceiver is used to send and receive messages, and the processor is used to read the computer program and execute the online education training camp in-class test question pushing method as described in the first aspect. For example, the memory may include, but is not limited to, a random access memory (RAM), a read-only memory (ROM), a flash memory (Flash Memory), a first-in-first-out memory (FIFO) and / or a first-in-last-out memory (FILO), etc.; the processor may include, but is not limited to, a microprocessor of the STM32F105 series. In addition, the computer device may also include, but is not limited to, a power module, a display screen and other necessary components.

[0097] The working process, working details and technical effects of the aforementioned computer device provided in the third aspect of this embodiment can be found in the method for pushing in-class test questions for the online education training camp described in the first aspect, and will not be described in detail here.

[0098] A fourth aspect of this embodiment provides a computer-readable storage medium storing instructions including the method for pushing in-class quiz questions for an online education training camp as described in the first aspect. Specifically, the computer-readable storage medium stores instructions that, when executed on a computer, execute the method for pushing in-class quiz questions for an online education training camp as described in the first aspect. The computer-readable storage medium refers to a data storage medium and may include, but is not limited to, a floppy disk, an optical disk, a hard disk, a flash memory, a USB flash drive, and / or a memory stick. The computer may be a general-purpose computer, a dedicated computer, a computer network, or other programmable device.

[0099] The working process, working details and technical effects of the aforementioned computer-readable storage medium provided in the fourth aspect of this embodiment can be referred to the method for pushing in-class test questions for the online education training camp described in the first aspect, and will not be repeated here.

[0100] A fifth aspect of this embodiment provides a computer program product, including a computer program or instructions, which, when executed by a computer, implements the method for pushing in-class quiz questions for an online education training camp as described in the first aspect. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device.

[0101] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention shall be included in the scope of protection of the present invention.

Claims

1. A method for pushing test questions for an online education training camp, characterized in that: include: Obtain the current and historical lecture notes for the online education training camp courses; Structural processing is performed on the current lecture text to obtain a current lecture structured text, and structural processing is also performed on the history lecture text to obtain a history lecture structured text, wherein the structural processing of the current lecture text to obtain the current lecture structured text specifically includes: using natural language processing technology to perform grammatical parsing processing on the current lecture text to obtain grammatical information including parts of speech and sentence structure; using rule / template matching to identify at least one phrase representing key knowledge points from the grammatical information; using a predefined knowledge system template / ontology to determine the key knowledge points represented by each phrase in the at least one phrase, and obtaining the current lecture structured text through structural representation; Performing knowledge point extraction processing on the structured text of the current class lecture notes to obtain a current class knowledge point set, and also performing knowledge point extraction processing on the structured text of the history class lecture notes to obtain a history class knowledge point set, wherein performing knowledge point extraction processing on the structured text of the current class lecture notes to obtain a current class knowledge point set specifically includes: applying the extractive question-answering model LangChain to extract a first knowledge point set from the structured text of the current class lecture notes, and applying ChatGPT, a first language model based on the Transformer structure and that has been pre-trained, to generate a second knowledge point set according to the structured text of the current class lecture notes; matching to obtain a knowledge point intersection of the first knowledge point set and the first knowledge point set; applying RoBERTa, a second language model pre-built based on natural language inference technology and that has been pre-trained, to determine whether each knowledge point in the knowledge point intersection has a semantically incoherent part; and using the knowledge point intersection from which the knowledge points with the semantically incoherent parts have been eliminated as the current class knowledge point set; The remaining set after removing the intersection of the current in-class knowledge point set and the historical in-class knowledge point set from the current in-class knowledge point set is used as the current in-class new knowledge point set; For each new knowledge point in the current set of new knowledge points learned in class, generate a corresponding plurality of in-class test questions; For each newly learned knowledge point, extract a segment of the target student's current in-class monitoring video, the segment of which is consistent with the display period of the handout page corresponding to the knowledge point, and import the segment of the in-class monitoring video into a pre-trained listening state recognition model based on a first artificial intelligence algorithm to output a first confidence level that the target student's listening state for the corresponding knowledge point is an attentive listening state; For each of the newly learned knowledge points, based on the corresponding first confidence level, the corresponding number of questions and / or the difficulty of the questions that are negatively correlated with the confidence level are determined, and based on the number of questions and / or the difficulty of the questions, at least one in-class test question is selected from the corresponding multiple in-class test questions and pushed to the target student.

2. The method for pushing test questions for online education training camp according to claim 1, characterized in that: For each new knowledge point in the current set of new knowledge points learned in class, a corresponding plurality of in-class test questions are generated, including: Using semantic matching technology to select a predefined question type template suitable for each new knowledge point in the set of new knowledge points to be learned in class, wherein the predefined question type template includes a question stem slot and an answer option slot, and the question types of the predefined question type template include multiple-choice questions, fill-in-the-blank questions, essay questions, true-or-false questions, and / or programming questions; For each newly learned knowledge point, the corresponding knowledge point content is inserted into the question stem slot of the corresponding predefined question template to obtain the corresponding question stem content, and then a language generation model based on the GPT series language model is applied to generate answer options that meet the requirements of both grammar and semantics based on the question stem content, and the answer options are inserted into the answer option slot of the corresponding predefined question template to obtain the corresponding answer content; For each newly learned knowledge point, assemble the corresponding question stem content and answer content to obtain the corresponding in-class test questions.

3. The method for pushing test questions for online education training camp according to claim 1, characterized in that: When there are multiple target students, for each newly learned knowledge point, at least one in-class test question is extracted from the corresponding multiple in-class test questions according to the corresponding question quantity and pushed to the target student, including: For each new knowledge point, traverse each target student in turn: according to the corresponding and determined number of questions for the current traversal student , randomly selected from the corresponding multiple in-class test questions unselected in-class test questions, and push the extraction results to the current traversal students, where Represents a positive integer.

4. The method for pushing test questions for online education training camp according to claim 1, characterized in that: The first artificial intelligence algorithm adopts a machine learning algorithm based on support vector machine, decision tree or random forest.

5. A device for pushing test questions for online education training camps, characterized in that: It includes a lecture text acquisition unit, a structured processing unit, a knowledge point extraction unit, a new knowledge point determination unit, a test question generation unit, a lecture status classification unit and a test question extraction unit which are sequentially connected in communication; The lecture text acquisition unit is used to acquire the current class lecture text and the historical class lecture text of the online education training camp teaching course; The structured processing unit is configured to perform structured processing on the current lecture text to obtain a structured text of the current lecture, and further perform structured processing on the history lecture text to obtain a structured text of the history lecture, wherein the structured processing on the current lecture text to obtain the structured text of the current lecture specifically includes: performing grammatical parsing processing on the current lecture text using natural language processing technology to obtain grammatical information including parts of speech and sentence structure; identifying at least one phrase representing a key knowledge point from the grammatical information using a rule / template matching method; determining the key knowledge point represented by each phrase in the at least one phrase using a predefined knowledge system template / ontology, and obtaining the structured text of the current lecture through structured representation; The knowledge point extraction unit is used to perform knowledge point extraction processing on the current class lecture structured text to obtain a current class knowledge point set, and also perform knowledge point extraction processing on the history class lecture structured text to obtain a history class knowledge point set, wherein the knowledge point extraction processing on the current class lecture structured text to obtain the current class knowledge point set specifically includes: applying the extractive question-answering model LangChain to extract a first knowledge point set from the current class lecture structured text, and applying the first language model ChatGPT based on the Transformer structure and pre-trained to generate a second knowledge point set according to the current class lecture structured text; matching to obtain a knowledge point intersection of the first knowledge point set and the first knowledge point set; applying the second language model RoBERTa pre-built based on natural language inference technology and pre-trained to determine whether each knowledge point in the knowledge point intersection has a semantically incoherent part; and using the knowledge point intersection from which the knowledge points with semantically incoherent parts have been eliminated as the current class knowledge point set; The newly learned knowledge point determination unit is configured to use the remaining set after removing the intersection of the current in-class knowledge point set and the historical in-class knowledge point set from the current in-class knowledge point set as the current in-class newly learned knowledge point set; The test question generating unit is used to generate a plurality of corresponding in-class test questions for each new knowledge point in the current set of new knowledge points learned in class; The listening state classification unit is configured to extract, from the target student's current in-class listening surveillance video, a segment of the in-class listening surveillance video whose acquisition period coincides with the display period of the handout page corresponding to the corresponding knowledge point, for each newly learned knowledge point, and import the segment of the in-class listening surveillance video into a pre-trained listening state recognition model based on a first artificial intelligence algorithm, and output a first confidence level that the target student's listening state at the corresponding knowledge point is an attentive listening state; The test question extraction unit is used to determine, for each of the newly learned knowledge points, a corresponding number of questions and / or a question difficulty that is negatively correlated with the confidence level according to the corresponding first confidence level, and extract at least one in-class test question from the corresponding multiple in-class test questions based on the number of questions and / or the question difficulty, and push the result to the target student.

6. A computer device, characterized in that: The method comprises a memory, a processor and a transceiver which are communicatively connected in sequence, wherein the memory is used to store a computer program, the transceiver is used to send and receive messages, and the processor is used to read the computer program and execute the method for pushing in-class test questions for an online education training camp as described in any one of claims 1 to 4.

7. A computer-readable storage medium, characterized in that The computer-readable storage medium stores instructions. When the instructions are executed on the computer, the method for pushing in-class test questions for an online education training camp as described in any one of claims 1 to 4 is executed.

8. A computer program product comprising a computer program or instructions, characterized in that When executed by a computer, the computer program or the instruction implements the method for pushing in-class test questions for an online education training camp as claimed in any one of claims 1 to 4.

Images