Learning condition comment generation method and device, related equipment and computer program product

Abstract

The invention discloses a learning condition comment generation method and device, related equipment and a computer program product. The method comprises the steps of obtaining a learning condition comment generation instruction; extracting entities of a set category from the learning condition comment generation instruction, wherein the set category at least comprises student identities; knowledge information matched with the entity is retrieved from a configured knowledge base, the knowledge information comprises retrieved target learning condition information and target external knowledge document fragments, and the knowledge base comprises learning condition information of students and external knowledge documents; determining external knowledge document fragments related to the target learning condition information from the target external knowledge document fragments; and taking the target learning condition information and the external knowledge document fragments related to the target learning condition information as background knowledge, injecting the background knowledge and a learning condition comment generation instruction into a prompt instruction, and sending the prompt instruction into the large model to obtain a learning condition comment generated by the large model. According to the scheme, the comment generation quality is improved.

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence technology, and more specifically, to a method, device, related equipment and computer program product for generating learning status comments. Background Art

[0002] With the development of artificial intelligence, more and more students choose online learning methods such as learning machine products. Similar products generally provide comment generation functions. However, the current comment generation methods basically use strong rules to match the constructed comment library, and the generated comments are basically stereotyped comment templates with low quality. Summary of the invention

[0003] In view of the above problems, this application is proposed to provide a method, device, related equipment and computer program product for generating learning status comments, so as to achieve the purpose of improving the quality of generated comments. The specific scheme is as follows:

[0004] In a first aspect, a method for generating a learning review is provided, comprising:

[0005] Get the instruction to generate learning status comments;

[0006] Extracting entities of a set category from the instruction for generating learning status comments, wherein the set category at least includes student identity;

[0007] Retrieving knowledge information matching the entity from a configured knowledge base, the knowledge information including the retrieved target learning information and target external knowledge document fragments, wherein the knowledge base includes the learning information of each student and the external knowledge document;

[0008] Determining, from the target external knowledge document fragments, an external knowledge document fragment related to the target learning situation information;

[0009] The target learning information and external knowledge document fragments related to the target learning information are used as background knowledge, and the background knowledge and the learning comment generation instruction are injected into the prompt instruction Prompt, and the Prompt is sent to the big model to obtain the learning comment generated by the big model.

[0010] In a possible design, in another implementation of the first aspect of the embodiment of the present application, the knowledge base further includes: classroom performance information and / or core learning literacy information of each student;

[0011] Then, the knowledge information matching the entity retrieved from the knowledge base also includes: the student identity corresponding to the student's target classroom performance and / or, target core learning literacy;

[0012] Then, the background knowledge also includes: the target classroom performance and / or the target core learning literacy.

[0013] In one possible design, in another implementation of the first aspect of the embodiment of the present application, the external knowledge document includes a knowledge point interpretation document and / or an examination syllabus, and the knowledge point interpretation document contains interpretation information of one or more knowledge points.

[0014] In a possible design, in another implementation of the first aspect of the embodiments of the present application, the knowledge base further includes a knowledge graph;

[0015] Before determining the reference document fragments related to the target learning situation information from the target external knowledge document fragments, the method further includes:

[0016] Retrieving target knowledge points related to the target learning situation information in the knowledge graph;

[0017] The process of determining the reference document fragment related to the target learning situation information from the target external knowledge document fragment includes:

[0018] An external knowledge document segment related to the target knowledge point is determined from the target external knowledge document segments.

[0019] In a possible design, in another implementation of the first aspect of the embodiment of the present application, the information in the knowledge base is pre-labeled; then the process of retrieving knowledge information matching the entity from the configured knowledge base includes:

[0020] The knowledge information corresponding to the label matching the entity is retrieved from the knowledge base.

[0021] In a possible design, in another implementation of the first aspect of the embodiment of the present application, the process of injecting the background knowledge and the learning situation comment generation instruction into the prompt instruction Prompt includes:

[0022] Obtain a prompt instruction Prompt format template, wherein the Prompt format template includes a task instruction, a background knowledge slot and an instruction slot, wherein the task instruction is used to instruct the large model to generate a learning status comment by combining the background knowledge in the background knowledge slot and the user instruction in the instruction slot;

[0023] The background knowledge is filled into the background knowledge slot, and the instruction for generating the learning situation comment is filled into the instruction slot to obtain a prompt instruction Prompt.

[0024] In a possible design, in another implementation of the first aspect of the embodiment of the present application, after determining the external knowledge document fragment related to the target learning situation information from the target external knowledge document fragment, it also includes:

[0025] De-duplication processing is performed on the external knowledge document fragments determined to be relevant to the target learning information.

[0026] In a possible design, in another implementation of the first aspect of the embodiment of the present application, the target learning situation information includes learning behavior information of different subjects and different time dimensions, and the learning behavior information includes at least one of the following:

[0027] Points of improvement, points of regression, weak points, accuracy rate of answers, and error rate of answers.

[0028] In a possible design, in another implementation of the first aspect of the embodiments of the present application, the setting category further includes at least one of the following:

[0029] Time, subject, region.

[0030] In a second aspect, a device for generating learning status comments is provided, comprising:

[0031] A user instruction acquisition unit, used to acquire instructions for generating learning status comments;

[0032] An entity extraction unit, used to extract entities of a set category from the instruction for generating learning status comments, wherein the set category at least includes student identity;

[0033] A first retrieval unit is used to retrieve knowledge information matching the entity from a configured knowledge base, wherein the knowledge information includes the retrieved target learning information and target external knowledge document fragments, wherein the knowledge base includes the learning information of each student and the external knowledge document;

[0034] A second retrieval unit is used to determine the external knowledge document fragments related to the target learning situation information from the target external knowledge document fragments;

[0035] The knowledge reorganization and comment generation unit is used to take the target learning information and external knowledge document fragments related to the target learning information as background knowledge, and inject the background knowledge and the learning comment generation instruction into the prompt instruction Prompt, send the Prompt into the big model, and obtain the learning comment generated by the big model.

[0036] In a third aspect, an electronic device is provided, comprising: a memory and a processor;

[0037] The memory is used to store programs;

[0038] The processor is used to execute the program to implement the learning status evaluation generation method described in any one of the first aspects of the present application.

[0039] In a fourth aspect, a readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the method for generating learning status comments described in any one of the first aspects of the present application is implemented.

[0040] In a fifth aspect, a computer program product is provided, comprising a computer program, which, when executed by a processor, implements the method for generating learning status comments described in any one of the first aspects of the present application.

[0041] By means of the above technical solution, the present application provides a personalized comment generation solution based on a large model. The present application can pre-configure a knowledge base, which includes the collected learning information of each student and the collected external knowledge documents. The external knowledge documents contain detailed introductions to different knowledge points. After obtaining the instruction for generating the learning comment, the instruction is firstly entity-extracted to obtain the entity extraction results of various categories set in advance, such as student identity information, etc., and then the knowledge information matching the extracted entity can be retrieved from the knowledge base, such as the target learning information matching the student identity retrieved, and the target external knowledge document fragment retrieved. Considering that the target external knowledge document fragment initially retrieved may belong to a relatively broad range, for this reason, the external knowledge document fragment related to the target learning information can be further retrieved from the target external knowledge document fragment, and the target learning information and the related external knowledge document fragment can be used as background knowledge, and the background knowledge and the instruction for generating the learning comment are injected into the Prompt, and the Prompt is sent to the large model, and the final learning comment is generated by the large model. Obviously, this application plan can combine students' learning information and matching external knowledge document fragments, and through the generation ability of a large model, generate personalized comments that are more closely related to students' learning situation, thereby improving the quality of the comments.

[0042] Furthermore, in the process of generating comments using students' learning information, the present application does not simply mechanically count and list the learning information, but further introduces external knowledge documents to match external knowledge document fragments related to students' learning information, and injects the learning information and external knowledge document fragments into the Prompt as background knowledge, providing the large model with richer reference information, thereby guiding the large model to generate higher quality learning comments. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Various other advantages and benefits will become apparent to those of ordinary skill in the art by reading the detailed description of the preferred embodiments below. The accompanying drawings are only for the purpose of illustrating the preferred embodiments and are not to be considered as limiting the present application. Also, the same reference symbols are used throughout the accompanying drawings to represent the same components. In the accompanying drawings:

[0044] Figure 1 A schematic diagram of an implementation system architecture of the learning status review generation method provided in an embodiment of the present application;

[0045] Figure 2 A schematic diagram of a terminal structure provided in an embodiment of the present application;

[0046] Figure 3 A schematic diagram of a server structure provided in an embodiment of the present application;

[0047] Figure 4 A schematic diagram of a process of generating a learning status review provided in an embodiment of the present application;

[0048] Figure 5 A schematic diagram of the composition of a knowledge base provided in an embodiment of the present application;

[0049] Figure 6 A schematic diagram of an example of a method for generating learning status comments provided in an embodiment of the present application;

[0050] Figure 7 A schematic diagram of the structure of a learning status review generation device provided in an embodiment of the present application;

[0051] Figure 8 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0052] Before introducing this application solution, the relevant concepts involved in this article are first explained:

[0053] prompt: Instructions. When interacting with AI (such as an artificial intelligence model), you need to send instructions to AI. It can be a text description, such as "Please recommend me a pop song" when you interact with AI, or it can be a parameter description in a certain format. For example, if you want AI to draw in a certain format, you need to describe the relevant drawing parameters.

[0054] Large models: In the field of artificial intelligence, large models usually refer to large-scale pre-trained models. Such models are called "large" because they can be pre-trained on a large amount of data and can be transferred to a variety of downstream tasks. Their full English name is Large Pre-Trained Models or Large-Scale Pre-Training Models. Large models are characterized by their large scale and contain billions or even more parameters, which help them learn complex patterns in the data. The emerging capabilities of large models include but are not limited to: contextual learning, instruction following, code generation, step-by-step reasoning capabilities, etc. Large models can include large language models (LLMs) and multimodal large models. Large language models are mainly used to process text modal data. Multimodal large models further integrate multimodal capabilities on the basis of large language models and can process information in multiple modalities, such as images, text, audio, etc.

[0055] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0056] It is understandable that before using the technical solutions disclosed in the embodiments of this application, the type, scope of use, usage scenarios, etc. of the personal information involved in this disclosure should be informed to the user and the user's authorization should be obtained in an appropriate manner in accordance with relevant laws and regulations.

[0057] This application provides a method for generating learning status comments, which can be applied to Figure 1 The system architecture shown in FIG. 1 may include a terminal 100 and a server 200. The server 200 may include one or more servers ( Figure 1 A server is included as an example for explanation).

[0058] The terminal 100 or the server 200 can be used alone to execute the learning situation review generation method provided in the embodiment of the present application. In addition, the terminal 100 and the server 200 can also be used in collaboration to execute the learning situation review generation method provided in the embodiment of the present application.

[0059] Next describe Figure 1 The product form of the mid-terminal 100;

[0060] The terminal 100 in the embodiment of the present application can be a mobile phone, a tablet computer, a learning machine, a teaching screen, a wearable device, a conference terminal, an augmented reality (AR) / virtual reality (VR) device, a laptop computer, an ultra-mobile personal computer (UMPC), a netbook, a personal digital assistant (PDA), etc., and the embodiment of the present application does not impose any restrictions on this.

[0061] Figure 2 An optional hardware structure diagram of the terminal 100 is shown.

[0062] refer to Figure 2 As shown, the terminal 100 may include a radio frequency unit 110, a memory 120, an input unit 130, a display unit 140, a camera 150 (optional), an audio circuit 160, a speaker 161, a microphone 162, an earphone jack 163 (optional), a processor 170, an external interface 180, a power supply 190 and other components. Those skilled in the art will appreciate that Figure 2 These are merely examples of terminals or multi-function devices and do not constitute limitations on the terminals or multi-function devices, which may include more or fewer components than those shown in the figures, or combinations of certain components, or different components.

[0063] The input unit 130 can be used to receive input digital or character information, and generate key signal input related to the user settings and function control of the portable multifunctional device. Specifically, the input unit 130 may include a touch screen 131 and / or other input devices 132. The touch screen 131 can collect the user's touch operations on or near it (such as the user's operation on or near the touch screen using any suitable object such as fingers, joints, stylus, etc.), and drive the corresponding connection device according to a pre-set program. The touch screen can detect the user's touch action on the touch screen, convert the touch action into a touch signal and send it to the processor 170, and can receive and execute the command sent by the processor 170; the touch signal at least includes the touch point coordinate information. The touch screen 131 can provide an input interface and an output interface between the terminal 100 and the user. In addition, the touch screen can be implemented using multiple types such as resistive, capacitive, infrared and surface acoustic wave. In addition to the touch screen 131, the input unit 130 can also include other input devices. Specifically, other input devices 132 may include, but are not limited to, one or more of a physical keyboard, function keys (such as a volume control key, a switch key, etc.), a trackball, a mouse, a joystick, and the like.

[0064] Among them, other input devices 132 can receive input data and so on.

[0065] The display unit 140 can be used to display information input by the user or information provided to the user, various menus of the terminal 100, interactive interfaces, file display and / or playback of any multimedia file. In the embodiment of the present application, the display unit 140 can be used to display various interactive interfaces, processing results, etc. in the learning status evaluation method.

[0066] The memory 120 can be used to store instructions and data. The memory 120 can mainly include an instruction storage area and a data storage area. The data storage area can store various data, such as multimedia files, texts, etc.; the instruction storage area can store software units such as operating systems, applications, instructions required for at least one function, or their subsets and extensions. It can also include a non-volatile random access memory; provide the processor 170 with hardware, software and data resources including management of computing and processing equipment, and support control software and applications. It is also used for the storage of multimedia files, and the storage of running programs and applications.

[0067] The processor 170 is the control center of the terminal 100. It uses various interfaces and lines to connect various parts of the entire terminal 100. By running or executing instructions stored in the memory 120 and calling data stored in the memory 120, it executes various functions of the terminal 100 and processes data, thereby controlling the terminal device as a whole. Optionally, the processor 170 may include one or more processing units; preferably, the processor 170 may integrate an application processor and a modem processor, wherein the application processor mainly processes the operating system, user interface and application program, and the modem processor mainly processes wireless communication. It is understandable that the above-mentioned modem processor may not be integrated into the processor 170. In some embodiments, the processor and the memory may be implemented on a single chip, and in some embodiments, they may also be implemented separately on separate chips. The processor 170 may also be used to generate corresponding operation control signals, send them to corresponding components of the computing and processing device, read and process data in the software, especially read and process data and programs in the memory 120, so that each functional module therein performs corresponding functions, thereby controlling the corresponding components to act according to the requirements of the instructions.

[0068] Among them, the memory 120 can be used to store software codes related to the learning situation comment generation method, the processor 170 can execute the steps of the learning situation comment generation method, and can also schedule other units (such as the above-mentioned input unit 130 and display unit 140) to realize corresponding functions.

[0069] The radio frequency unit 110 (optional) can be used for receiving and sending information or receiving and sending signals during a call, for example, after receiving the downlink information of the base station, it is sent to the processor 170 for processing; in addition, the designed uplink data is sent to the base station. Generally, the RF circuit includes but is not limited to an antenna, at least one amplifier, a transceiver, a coupler, a low noise amplifier (Low Noise Amplifier, LNA), a duplexer, etc. In addition, the radio frequency unit 110 can also communicate with network devices and other devices through wireless communication. The wireless communication can use any communication standard or protocol, including but not limited to Global System of Mobile communication (Global System of Mobile communication, GSM), General Packet Radio Service (General Packet Radio Service, GPRS), Code Division Multiple Access (Code Division Multiple Access, CDMA), Wideband Code Division Multiple Access (Wideband Code Division Multiple Access, WCDMA), Long Term Evolution (Long Term Evolution, LTE), email, Short Messaging Service (SMS), etc.

[0070] In the embodiment of the present application, the RF unit 110 can send data to the server 200 and receive the processing result sent by the server 200. Exemplarily, the RF unit 110 sends the received user's learning review generation instruction to the server 200, and the server 200 obtains the learning review generated by the large model and returns the learning review to the terminal 100 for output.

[0071] It should be understood that the radio frequency unit 110 is optional and can be replaced by other communication interfaces, such as a network port.

[0072] The terminal 100 also includes a power supply 190 (such as a battery) for supplying power to various components. Preferably, the power supply can be logically connected to the processor 170 through a power management system, so that the power management system can manage functions such as charging, discharging, and power consumption.

[0073] The terminal 100 further includes an external interface 180 , which may be a standard Micro USB interface or a multi-pin connector, and may be used to connect the terminal 100 to communicate with other devices, or to connect a charger to charge the terminal 100 .

[0074] Although not shown, the terminal 100 may also include a flashlight, a wireless fidelity (WiFi) module, a Bluetooth module, sensors with different functions, etc., which are not described in detail here. Some or all of the methods described below may be applied in the following embodiments. Figure 2 In the terminal 100 shown.

[0075] Next describe Figure 1 The product form of the server 200;

[0076] Figure 3 A structural diagram of a server 200 is provided, such as Figure 3 As shown, the server 200 includes a bus 201, a processor 202, a communication interface 203, and a memory 204. The processor 202, the memory 204, and the communication interface 203 communicate with each other via the bus 201.

[0077] The bus 201 may be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus. The bus may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 3 Only one thick line is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0078] The processor 202 may be any one or more of a central processing unit (CPU), a graphics processing unit (GPU), a microprocessor (MP), or a digital signal processor (DSP).

[0079] The memory 204 may include a volatile memory, such as a random access memory (RAM). The memory 204 may also include a non-volatile memory, such as a read-only memory (ROM), a flash memory, a hard disk drive (HDD), or a solid state drive (SSD).

[0080] Among them, the memory 204 can be used to store software codes related to the method for generating learning status comments, the processor 202 can execute the steps of the method for generating learning status comments, and can also schedule other units to realize corresponding functions.

[0081] It should be understood that the above-mentioned terminal 100 and server 200 can be centralized or distributed devices, and the processors in the above-mentioned terminal 100 and server 200 (such as processor 170 and processor 202) can be hardware circuits (such as application specific integrated circuit (ASIC), field-programmable gate array (FPGA), general-purpose processor, digital signal processor (DSP), microprocessor or microcontroller, etc.), or a combination of these hardware circuits. For example, the processor can be a hardware system with an instruction execution function, such as a CPU, DSP, etc., or a hardware system without an instruction execution function, such as an ASIC, FPGA, etc., or a combination of the above-mentioned hardware systems without an instruction execution function and hardware systems with an instruction execution function.

[0082] At this stage, the mainstream learning evaluation schemes are mainly divided into three categories. One is to manually pre-select student portraits in several dimensions, such as gender, personality, position, student grades, or add user-specified keywords and other information. The system then matches and retrieves the established evaluation template library based on the content of the portrait, and randomly gives template comments. If the user is not satisfied, he or she can replace it by himself or herself. The replacement method generally adopts a random or similarity-based method to meet the needs. The other type is a simple data statistics of the student's learning records, such as the number of answers, answering time, answering accuracy, and learning time, which are only summarized in the form of charts. The last type uses NLP-related deep learning technology, which requires professional teaching and research teachers to annotate the corresponding learning evaluation according to the student's learning behavior, and then use the annotated data to train a supervised evaluation generation model.

[0083] The existing mainstream comment schemes have several shortcomings: the first type is based on matching and searching by rules and template libraries. The comments given are relatively simple and templated, which cannot meet the requirements of personalized comments. The second type of scheme mechanically counts and displays students' academic record indicators. The presentation format is fixed and lacks in-depth analysis of knowledge. The third type of scheme uses supervised deep neural networks, which often have good generalization performance in feature prediction, but require expensive teacher labeling costs. In addition, due to the differences in educational regions, frequent changes in lesson plans or knowledge points also greatly reduce the generalization performance of deep learning models.

[0084] The present application provides a method for generating a learning status review, which is applied to Figure 1 Taking the computer device in the example as an example, the computer device can be Figure 1 The terminal 100 or the system consisting of the terminal 100 and the server 200. Figure 4 The learning situation evaluation generation method specifically includes the following steps:

[0085] Step S100, obtaining instructions for generating learning status comments.

[0086] Specifically, the instruction for generating the learning evaluation can be input by an external user. For example, a teacher inputs the following instruction: "Help me generate a mathematics evaluation of Wang xx in Class 3, Grade 9 of xx Middle School in xx area, Anhui Province in the past six months."

[0087] In addition, the instruction for generating learning comments can also be an instruction sent by other functional modules in the learning platform to the learning comments generation module. For example, when the homework correction module determines that the comment generation trigger conditions are met (such as the user clicks on the comment generation control, etc.), it automatically extracts relevant information to generate the instruction for generating learning comments and sends it to the learning comments generation module.

[0088] The instruction for generating academic evaluation generally includes some entity information, such as the student's identity information (student name, student number, school, age, class, etc.), region information, time information, subject information, etc.

[0089] Of course, the types of entities contained in the instruction for generating learning status comments obtained in different scenarios may not be exactly the same, and the entity information may include one or more combinations of the above examples. Considering that in order to facilitate the subsequent accurate generation of learning status comments for students, the instruction for generating learning status comments obtained in this step may carry at least the student identity information.

[0090] Step S110: extracting entities of set categories from the instruction for generating learning status comments, wherein the set categories at least include student identities.

[0091] Specifically, the present application may predefine the slot entity category to be extracted, and the slot entity category may at least include student identity, such as student name, student number, school, age, class, etc.

[0092] Further optionally, the slot entity category may also include one or more of the following examples: time, subject, region.

[0093] Take the query for generating learning evaluation in the previous step as an example: "Help me generate a math evaluation of Wang XX in Class 3, Grade 9 of XX Middle School in XX Region, Anhui Province in the past six months." as an example. The entities that can be extracted include:

[0094] Region: Anhui xx area;

[0095] School: xx Middle School;

[0096] Name: Wang xx;

[0097] Grade: Junior 3;

[0098] Class: Class 3;

[0099] Subject: Mathematics;

[0100] Time: Nearly half a year.

[0101] In some possible implementations, an entity extraction model may be used in this step to extract entities of a set category from the instruction for generating learning status comments. In addition, a rule-based approach may be used to match entities of a set category from the instruction for generating learning status comments, or a large model may be used to implement the entity extraction process, etc.

[0102] Step S120: Retrieve knowledge information matching the entity from the configured knowledge base, wherein the knowledge information includes the retrieved target learning situation information and target external knowledge document fragments.

[0103] The knowledge base includes the learning information of each student and external knowledge documents. The knowledge base can be created in advance before executing the learning review generation task.

[0104] The knowledge base created in this application may at least include the learning information of different students, as well as external knowledge documents.

[0105] In a possible implementation, the student's learning information may include learning behavior information of different subjects and different time dimensions, and the learning behavior information includes but is not limited to:

[0106] Points of improvement, points of regression, weak points, accuracy rate of answers, error rate of answers, etc.

[0107] Among them, the progress points are used to compare the initial students' mastery of test points, record the test points with significant improvement in mastery, and use them to encourage students.

[0108] Regression points: used to compare the initial students’ mastery of test points, record the test points where the mastery has declined significantly, and remind students to pay attention.

[0109] Weak points: record the students’ mastery of test points after assessment, measure the students’ true mastery of a certain test point, and identify the test points with lower mastery as weak points.

[0110] The above only illustrates several optional examples of learning behavior information, where different learning behavior information can be obtained through rules or preconfigured models.

[0111] External knowledge documents contain detailed descriptions of different knowledge points. External knowledge documents can be in various forms, including but not limited to: exam syllabus, knowledge point interpretation documents, etc. Among them, the exam syllabus can describe in detail the knowledge points, skills or concepts that need to be mastered in each subject or module, as well as a general description of the difficulty of the exam, such as the proportion of basic questions, medium questions, and difficult questions. The exam syllabus can help determine the weak points of the exam that students need to pay attention to. For example, a student has 5 weak points, but according to the exam syllabus, 2 of them are not tested in the area, so according to the exam syllabus, students can be reminded to pay attention to the other 3 points.

[0112] The exam syllabus may change for different regions and time dimensions. Therefore, the knowledge base may include exam syllabi for different regions and time dimensions.

[0113] For the knowledge point interpretation document, it can include detailed interpretation content of each knowledge point, such as interpretation of knowledge points that are easy to make mistakes or knowledge points with higher difficulty, etc. The knowledge point interpretation document can be a knowledge point interpretation document with a unified configuration, that is, it is suitable for teaching needs in different regions and different time dimensions. In addition, according to the needs of actual teaching scenarios, different knowledge point interpretation documents can also be set for different regions and different time dimensions.

[0114] In this step, for each entity extracted in the previous step, the knowledge information matching the entity can be retrieved in the knowledge base for each entity or a combination of two or more entities. The retrieved knowledge information can include target academic information matching the student identity, and target external knowledge documents retrieved from the external knowledge documents.

[0115] For example, the extracted entities include the student's name, school, age, class, and region. The student's name, school, age, and class information can be used to retrieve the target learning information of the current student. The student's region information can be used to retrieve matching syllabus and knowledge point interpretation document fragments as target external knowledge documents.

[0116] In some possible implementations, combined with Figure 5 As shown, the knowledge base of the present application may further include at least one or more of the following:

[0117] Each student’s classroom performance information, core learning literacy information, etc.

[0118] Among them, classroom performance information: teachers can give certain students relevant classroom performance labels during normal classes, such as lively and positive, strong sense of responsibility, etc.

[0119] Core learning literacy information: This part of information can use big data to analyze students' answer records, answering time, test points, etc., and abstract different types of learning literacy, such as learning attitude, learning willpower, learning efficiency, learning goals, etc. Taking learning attitude as an example, students' learning attitude can be analyzed through information such as the time they take to answer and their handwriting.

[0120] The core learning literacy of the above example can be determined by a corresponding algorithm model, which will not be described in detail in this embodiment.

[0121] Based on the classroom performance information and core learning literacy information of each student included in the knowledge base, the knowledge information matching the entity retrieved from the knowledge base in this step may also include: the target classroom performance of the student corresponding to the student identity, and / or, target core learning literacy.

[0122] For example, the student identity information may be used as a search condition to search the knowledge base for the classroom performance information and core learning literacy information of the student corresponding to the current student identity.

[0123] Step S130: Determine the external knowledge document fragments related to the target learning situation information from the target external knowledge document fragments.

[0124] Specifically, in the previous step, based on the extracted entities, the target learning information and target external knowledge document fragments are preliminarily retrieved from the knowledge base.

[0125] Considering that the target external knowledge document fragment is retrieved based on the extracted entity, for example, based on the extracted regional type and time type entities, the matching target external knowledge document fragment is retrieved, such as the test outline fragment and the knowledge point interpretation document fragment. The target external knowledge document fragment may cover a wide range. In order to further narrow the scope of external knowledge, in this step, the target learning information can be used as the retrieval condition to further retrieve the external knowledge document fragment related to the target learning information from the target external knowledge document fragment.

[0126] In one possible implementation, combining Figure 5 As shown, the knowledge base in this application can further include a knowledge graph, which provides multi-dimensional information related to test points to help find students' real weaknesses. Such as the relationship between the predecessors and successors of the test points, the attributes of the test points themselves (such as difficulty, test frequency, question types, etc.), and the traceability analysis of the test points. In addition, the knowledge graph itself can also take into account the regionality of the region, such as configuring corresponding knowledge graphs for different regions.

[0127] On this basis, the present application can retrieve target knowledge points related to the target learning information from the knowledge graph. Taking the deficiencies included in the target learning information as an example, the target knowledge points related to the deficiencies can be retrieved from the knowledge graph, thereby finding the students' real weak knowledge points.

[0128] Furthermore, external knowledge document fragments related to the target knowledge point may be retrieved from the target external knowledge document.

[0129] It should be noted here that for each type of learning behavior information included in the target learning information, such as improvement points, regression points, weaknesses, correct answer rate, wrong answer rate, etc., the external knowledge document fragments related to each learning behavior information can be retrieved in the target external knowledge document. Alternatively, the relevant external knowledge document fragments can be retrieved in the external knowledge document for one or several types of configured learning behavior information.

[0130] In this step, external knowledge document fragments related to the target learning information can be retrieved from the target external knowledge document for the second time, thereby narrowing the scope of the recalled external knowledge and obtaining external knowledge that is more relevant to the target learning information.

[0131] In a possible implementation, after obtaining the external knowledge document fragments related to the target learning information, this step can perform deduplication processing on them to remove repeated fragments.

[0132] Step S140: Use the target learning information and external knowledge document fragments related to the target learning information as background knowledge, inject the background knowledge and the learning comment generation instruction into the prompt instruction Prompt, send the Prompt to the big model, and obtain the learning comment generated by the big model.

[0133] Specifically, after the above steps, at least two types of information can be retrieved from the knowledge base, namely, the target learning information and the external knowledge document fragments related to the target learning information. In this step, the two types of information can be combined as background knowledge, and then the background knowledge and the learning comment generation instruction can be injected into the prompt instruction Prompt, and the Prompt can be used to guide the large model to generate learning comments.

[0134] In a possible implementation, this embodiment provides a prompt instruction Prompt format template. The template includes a task instruction, a background knowledge slot and an instruction slot. Among them, the task instruction is used to instruct the large model to combine the background knowledge in the background knowledge slot and the user instruction in the instruction slot to generate a learning status comment.

[0135] Based on the above-mentioned Prompt format template, this step can fill the acquired background knowledge into the background knowledge slot, fill the learning situation comment generation instruction into the instruction slot, and obtain the prompt instruction Prompt.

[0136] In this step, with the help of the powerful context understanding and text generation capabilities of the large language model, personalized learning comments that better match the student's learning information can be generated under the instruction of the prompt. In addition, since the background knowledge contained in the prompt not only contains the target learning information, but also contains relevant external knowledge document fragments, the external knowledge document fragments can fully interpret the relevant knowledge, thereby assisting the large model to generate higher quality learning comments.

[0137] In some embodiments of the present application, for the constructed knowledge base, in order to facilitate the retrieval of location information from the knowledge base, the present application may pre-process the knowledge base.

[0138] The preprocessing process includes but is not limited to: labeling and vectorizing the knowledge information in the knowledge base.

[0139] Labeling process:

[0140] The learning information, classroom performance information and core learning literacy information in the knowledge base can be labeled by name, time, subject, etc.

[0141] The examination syllabus in the knowledge base can be labeled using region, time, subject, etc.

[0142] For knowledge point interpretation documents in the knowledge base, they can be labeled using the knowledge point names.

[0143] For the knowledge base after labeling, the process of retrieving knowledge information matching the entity from the knowledge base in the above steps may specifically include:

[0144] Retrieve knowledge information corresponding to the label matching the entity in the knowledge base.

[0145] By labeling the knowledge base, it is easier to search the knowledge base and improve the search efficiency.

[0146] Vectorization process:

[0147] Here, the process of vectorization processing can be for various types of knowledge in the knowledge base, or for some types of knowledge in the knowledge base. In this embodiment, only the process of vectorization processing of the examination syllabus is used as an example for explanation.

[0148] In order to find the context fragments in the examination syllabus more accurately and considering the length limit of the context window input by the large model, in this embodiment, the structured examination syllabus can be divided into a certain number of chunks.

[0149] There are many ways to divide the exam outline, for example:

[0150] (1) It can be segmented directly according to punctuation marks and number of characters, which is simple and efficient.

[0151] (2) A sliding window can be used to slide the context to generate overlapping blocks of a certain size, so that each block has a certain amount of overlapping information.

[0152] Furthermore, the divided chunks can be vectorized separately to obtain a vector representation of each chunk.

[0153] Based on the above vectorization processing, the subsequent retrieval process can be based on the retrieval method of vector similarity.

[0154] The learning review generation method of the present application can be applied to a variety of scenarios where learning reviews need to be generated. Taking the scenario of a class or a large-scale online class-like group as an example, the teacher can propose a learning review generation instruction in a natural language way, and then generate learning reviews through a large model, which can help the teacher quickly know the current status of the comments of any student in the class or group.

[0155] Combination Figure 6 , this embodiment introduces an example of a learning evaluation generation process.

[0156] Assume that the comment generation method is applied to a large collective class of online education. The query input by the teacher for generating the learning situation comment can be: "Help me generate a math comment for Wang XX in Class 3, Grade 9 of XX Middle School in XX area, Anhui Province in the past six months."

[0157] The query first extracts each entity through the NER named entity model or the big model, and then searches for matching knowledge information in the built knowledge base based on the extracted entities. The extracted entities are as follows: {"area":"Anhui","school":"xxx Middle School","name":"Wang xx","subject":"math","phase":"junior high school","class":"Class 3","time":"within half a year before the current time"}.

[0158] On the one hand, through recall strategy 1, for name, school class and time, the corresponding target learning information, classroom performance and core learning literacy of the student can be retrieved through the learning information, core learning literacy and classroom performance based on the knowledge graph. Among them, examples of target learning information include progress points, shortcomings, test question correct answer rate, etc. On the other hand, through recall strategy 2, for time and region, the corresponding document block (corresponding Figure 6 ), and then retrieve documents related to the target learning information from the pre-selected documents according to the target learning information retrieved above. For example, a retrieval method based on vector similarity is used to retrieve the TopK documents with the highest relevance to the target learning information from the pre-selected documents, and perform deduplication and sorting.

[0159] The following is an example of the application of the classroom performance, core learning literacy competencies, target learning information study, knowledge point interpretation document fragment knowledge, and exam outline fragment exam_point obtained through the above two recall strategies:

[0160] {

[0161] "performance": "Study hard, listen attentively in class, and complete homework with high quality, xxx.",

[0162] "competencies": "Have independent thinking ability, can analyze problems from multiple angles and dialectically; profound cultural background, xxx.",

[0163] "study": {

[0164] "Progress points": "Knowledge point 1, knowledge point 3, knowledge point 4,...",

[0165] "Deficiencies": "Knowledge point 2, knowledge point 5, knowledge point 6",

[0166] "Easy to make mistakes": "Knowledge point 7,...",

[0167] "Correct answer rate":"xxx"

[0168] },

[0169] "knowledge": "Knowledge point 5 is mainly based on xxx, the easy-to-make mistakes are xxx, the difficult points are xxx, and the xxx method is generally used; the key to knowledge point 7 is to clarify the concept of xxx, which needs to be discussed in categories according to specific situations, with special attention paid to the key points of xxx.",

[0170] "exam_point": [

[0171] {

[0172] "Segment 1": "Knowledge point 1: Basic content, requiring candidates to master it proficiently and correctly answer simple questions involving this knowledge point in the exam."

[0173] },

[0174] {

[0175] "Clip 2": "Knowledge point 3: In the extension of basic knowledge, candidates need to understand and apply relevant concepts, especially showing flexibility when combining them with real-life situations."

[0176] },

[0177] {

[0178] "Segment 3": "Knowledge point 2: One of the core points of the exam, requiring candidates to have a deep understanding and solve medium and above difficulty problems involving this knowledge point."

[0179] },

[0180] {

[0181] "Chapter 4": "Knowledge point 4: The key point of theoretical knowledge. Candidates need to fully grasp the relevant principles and be able to demonstrate high analytical and application capabilities in comprehensive application questions."

[0182] },

[0183] {

[0184] "Segment 5": "Knowledge point 5: This is difficult knowledge, involving complex and comprehensive problems. The test requires the examinee to understand the application of relevant concepts in practice."

[0185] },

[0186] {

[0187] "Chapter 6": "Knowledge point 6: It requires candidates to have strong reasoning and calculation abilities. It is an important difficulty in the exam and candidates should pay special attention to it."

[0188] },

[0189] {

[0190] "Segment 7": "Knowledge point 7: This part is prone to errors and is a basic point. Candidates should pay special attention to the accurate understanding and application of concepts in preparation to avoid making common mistakes in the exam."

[0191] }, ... ]

[0194] }.

[0195] Finally, the knowledge Context0 obtained by recall strategy 1 and the knowledge Context1 obtained by recall strategy 2 are merged and Concat is performed to obtain the merged background knowledge Context.

[0196] The background knowledge Context and the user's query are injected into the Prompt, and then the Prompt is sent to the big model for reasoning to generate learning comments. The Prompt example is as follows:

[0197] "[Task Description]

[0198] You are a middle school teacher with rich teaching experience. You know how to discover and solve the weaknesses in students' learning. Please refer to [Background Knowledge] and [Instructions] and give a detailed comment.

[0199]

Background knowledge

[0200] {Content}

[0201]

instruction

[0202] Help me generate a math review for Wang xx from Class 3, Junior 3, xx Middle School, xx District, Anhui Province in the past six months. "

[0203] The following is a description of the learning situation comment generation device provided in an embodiment of the present application. The learning situation comment generation device described below and the learning situation comment generation method described above can be referenced to each other.

[0204] See also Figure 7 , Figure 7 This is a schematic diagram of the structure of a learning status comment generation device disclosed in an embodiment of the present application.

[0205] like Figure 7 As shown, the device may include:

[0206] A user instruction acquisition unit 11 is used to acquire an instruction for generating a learning status comment;

[0207] An entity extraction unit 12, used to extract entities of a set category from the instruction for generating learning status comments, wherein the set category at least includes student identity;

[0208] A first retrieval unit 13 is used to retrieve knowledge information matching the entity from a configured knowledge base, the knowledge information including the retrieved target learning information and target external knowledge document fragments, wherein the knowledge base includes the learning information of each student and the external knowledge document;

[0209] A second retrieval unit 14 is used to determine the external knowledge document fragments related to the target learning situation information from the target external knowledge document fragments;

[0210] The knowledge reorganization and comment generation unit 15 is used to take the target learning information and the external knowledge document fragments related to the target learning information as background knowledge, and inject the background knowledge and the learning comment generation instruction into the prompt instruction Prompt, send the Prompt to the big model, and obtain the learning comment generated by the big model.

[0211] In a possible implementation, the setting category further includes at least one of the following:

[0212] Time, subject, region.

[0213] In one possible implementation, the knowledge base also includes: classroom performance information and / or core learning literacy information of each student; the knowledge information matching the entity retrieved from the knowledge base by the first retrieval unit also includes: the target classroom performance and / or target core learning literacy of the student corresponding to the student identity; the background knowledge obtained by the knowledge reorganization and comment generation unit also includes: the target classroom performance and / or the target core learning literacy.

[0214] In a possible implementation, the external knowledge document includes a knowledge point interpretation document and / or an examination syllabus, and the knowledge point interpretation document contains interpretation information of more than one knowledge point.

[0215] In a possible implementation, the knowledge base also includes a knowledge graph; then the device of the present application may also include: a knowledge graph retrieval unit, which is used to retrieve target knowledge points related to the target learning information in the knowledge graph before the second retrieval unit determines the reference document fragment related to the target learning information from the target external knowledge document fragment. Based on this, the process of the second retrieval unit determining the reference document fragment related to the target learning information from the target external knowledge document fragment includes:

[0216] An external knowledge document segment related to the target knowledge point is determined from the target external knowledge document segments.

[0217] In a possible implementation, the information in the knowledge base is pre-labeled; then the process of the first retrieval unit retrieving the knowledge information matching the entity from the configured knowledge base includes:

[0218] The knowledge information corresponding to the label matching the entity is retrieved from the knowledge base.

[0219] In a possible implementation, the process in which the knowledge reorganization and comment generation unit injects the background knowledge and the learning situation comment generation instruction into the prompt instruction Prompt includes:

[0220] Obtain a prompt instruction Prompt format template, wherein the Prompt format template includes a task instruction, a background knowledge slot and an instruction slot, wherein the task instruction is used to instruct the large model to generate a learning status comment by combining the background knowledge in the background knowledge slot and the user instruction in the instruction slot;

[0221] The background knowledge is filled into the background knowledge slot, and the instruction for generating the learning situation comment is filled into the instruction slot to obtain a prompt instruction Prompt.

[0222] In a possible implementation, the device of the present application may further include:

[0223] A deduplication processing unit is used to deduplicate the external knowledge document fragments related to the target learning information obtained by the second retrieval unit.

[0224] In a possible implementation, the target learning situation information includes learning behavior information of different subjects and different time dimensions, and the learning behavior information includes at least one of the following:

[0225] Points of improvement, points of regression, weak points, accuracy rate of answers, and error rate of answers.

[0226] The present application also provides an electronic device in an embodiment. Figure 8 As shown, it shows a schematic diagram of the structure of an electronic device suitable for implementing the embodiment of the present application. The electronic device in the embodiment of the present application may include but is not limited to fixed terminals such as mobile phones, tablet computers, learning machines, teaching large screens, wearable devices, etc. Figure 8 The electronic device shown is merely an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0227] like Figure 8 As shown, the electronic device may include a processing device (such as a central processing unit, a graphics processing unit, etc.) 601, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 602 or a program loaded from a storage device 608 to a random access memory (RAM) 603 to implement the learning situation comment generation method of the aforementioned embodiment of the present application. When the electronic device is powered on, various programs and data required for the operation of the electronic device are also stored in RAM 603. The processing device 601, ROM 602, and RAM 603 are connected to each other via a bus 604. An input / output (I / O) interface 605 is also connected to the bus 604.

[0228] Typically, the following devices may be connected to the I / O interface 605: an input device 606 including, for example, a touch screen, a touch pad, a keyboard, a mouse, a camera, a microphone, an accelerometer, a gyroscope, etc.; an output device 607 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; a storage device 608 including, for example, a memory card, a hard disk, etc.; and a communication device 609. The communication device 609 may allow the electronic device to communicate with other devices wirelessly or by wire to exchange data. Although Figure 8 An electronic device having various devices is shown, but it should be understood that it is not required to implement or possess all the devices shown. More or fewer devices may be implemented or possessed instead.

[0229] Also provided in an embodiment of the present application is a computer program product, including computer-readable instructions. When the computer-readable instructions are executed on an electronic device, the electronic device implements any one of the learning status evaluation generation methods provided in the embodiments of the present application.

[0230] A computer-readable storage medium is also provided in an embodiment of the present application. The storage medium carries one or more computer programs. When one or more computer programs are executed by an electronic device, the electronic device can implement any one of the learning status evaluation generation methods provided in the embodiment of the present application.

[0231] It should also be noted that the device embodiments described above are merely schematic, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed over multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. In addition, in the drawings of the device embodiments provided by the present application, the connection relationship between the modules indicates that there is a communication connection between them, which may be specifically implemented as one or more communication buses or signal lines.

[0232] Through the description of the above implementation mode, the technicians in the field can clearly understand that the present application can be implemented by means of software plus necessary general hardware, and of course, it can also be implemented by special hardware including special integrated circuits, special CPUs, special memories, special components, etc. In general, all functions completed by computer programs can be easily implemented by corresponding hardware, and the specific hardware structure used to implement the same function can also be various, such as analog circuits, digital circuits or special circuits. However, for the present application, software program implementation is a better implementation mode in more cases. Based on such an understanding, the technical solution of the present application is essentially or the part that contributes to the prior art can be embodied in the form of a software product, which is stored in a readable storage medium, such as a computer floppy disk, a U disk, a mobile hard disk, a ROM, a RAM, a disk or an optical disk, etc., including a number of instructions to enable a computer device (which can be a personal computer, a training device, or a network device, etc.) to execute the methods described in each embodiment of the present application.

[0233] In the above embodiments, all or part of the embodiments may be implemented by software, hardware, firmware or any combination thereof. When implemented by software, all or part of the embodiments may be implemented in the form of a computer program product.

[0234] The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions may be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions may be transmitted from a website site, a computer, a training device, or a data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) mode to another website site, computer, training device, or data center. The computer-readable storage medium may be any available medium that a computer can store or a data storage device such as a training device, a data center, etc. that includes one or more available media integrations. The available medium may be a magnetic medium, (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid-state drive (SSD)), etc.

[0235] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The various embodiments can be combined as needed, and the same or similar parts can refer to each other.

Claims

1. A method for generating learning status comments, characterized in that: include: Get the instruction to generate learning status comments; Extracting entities of a set category from the instruction for generating learning status comments, wherein the set category at least includes student identity; Retrieving knowledge information matching the entity from a configured knowledge base, the knowledge information including the retrieved target learning information and target external knowledge document fragments, wherein the knowledge base includes the learning information of each student and the external knowledge document; Determining, from the target external knowledge document fragments, an external knowledge document fragment related to the target learning situation information; The target learning information and external knowledge document fragments related to the target learning information are used as background knowledge, and the background knowledge and the learning comment generation instruction are injected into the prompt instruction Prompt, and the Prompt is sent to the big model to obtain the learning comment generated by the big model.

2. The method according to claim 1, characterized in that The knowledge base also includes: classroom performance information and / or core learning literacy information of each student; Then, the knowledge information matching the entity retrieved from the knowledge base also includes: the student identity corresponding to the student's target classroom performance and / or, target core learning literacy; Then, the background knowledge also includes: the target classroom performance and / or the target core learning literacy.

3. The method according to claim 1, characterized in that The external knowledge document includes a knowledge point interpretation document and / or an examination syllabus, and the knowledge point interpretation document contains interpretation information of more than one knowledge point.

4. The method according to claim 1, characterized in that The knowledge base also includes a knowledge graph; Before determining the reference document fragments related to the target learning situation information from the target external knowledge document fragments, the method further includes: Retrieving target knowledge points related to the target learning situation information in the knowledge graph; The process of determining the reference document fragment related to the target learning situation information from the target external knowledge document fragment includes: An external knowledge document segment related to the target knowledge point is determined from the target external knowledge document segments.

5. The method according to claim 1, characterized in that The information in the knowledge base is pre-labeled; the process of retrieving knowledge information matching the entity from the configured knowledge base includes: The knowledge information corresponding to the label matching the entity is retrieved from the knowledge base.

6. The method according to claim 1, characterized in that The process of injecting the background knowledge and the learning situation comment generation instruction into the prompt instruction Prompt includes: Obtain a prompt instruction Prompt format template, wherein the Prompt format template includes a task instruction, a background knowledge slot and an instruction slot, wherein the task instruction is used to instruct the large model to generate a learning status comment by combining the background knowledge in the background knowledge slot and the user instruction in the instruction slot; The background knowledge is filled into the background knowledge slot, and the instruction for generating the learning situation comment is filled into the instruction slot to obtain a prompt instruction Prompt.

7. The method according to claim 1, characterized in that After determining the external knowledge document fragments related to the target learning situation information from the target external knowledge document fragments, the method further includes: De-duplication processing is performed on the external knowledge document fragments determined to be relevant to the target learning information.

8. The method according to any one of claims 1 to 7, characterized in that: The target learning situation information includes learning behavior information of different subjects and different time dimensions, and the learning behavior information includes at least one of the following: Points of improvement, points of regression, weak points, accuracy rate of answers, and error rate of answers.

9. The method according to any one of claims 1 to 7, characterized in that: The setting category also includes at least one of the following: Time, subject, region.

10. A device for generating learning status comments, characterized in that: include: A user instruction acquisition unit, used to acquire instructions for generating learning status comments; An entity extraction unit, used to extract entities of a set category from the instruction for generating learning status comments, wherein the set category at least includes student identity; A first retrieval unit is used to retrieve knowledge information matching the entity from a configured knowledge base, wherein the knowledge information includes the retrieved target learning information and target external knowledge document fragments, wherein the knowledge base includes the learning information of each student and the external knowledge document; A second retrieval unit is used to determine the external knowledge document fragments related to the target learning situation information from the target external knowledge document fragments; The knowledge reorganization and comment generation unit is used to take the target learning information and external knowledge document fragments related to the target learning information as background knowledge, and inject the background knowledge and the learning comment generation instruction into the prompt instruction Prompt, send the Prompt into the big model, and obtain the learning comment generated by the big model.

11. An electronic device, characterized in that: include: Memory and processor; The memory is used to store programs; The processor is used to execute the program to implement the various steps of the method for generating learning status comments as described in any one of claims 1 to 9.

12. A readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, each step of the method for generating learning status comments as described in any one of claims 1 to 9 is implemented.

13. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, each step of the method for generating learning status comments as described in any one of claims 1 to 9 is implemented.

Images