Graphical programming homework correction method and device based on artificial intelligence

Abstract

The invention discloses a graphical programming homework correction method and device based on artificial intelligence. The method comprises the following steps: acquiring a graphical programming job, and converting the graphical programming job into a graphical programming job in an analyzable format; and based on a pre-constructed knowledge base, utilizing the AI large model to carry out correction analysis on the graphical programming job with the analyzable format to obtain a correction analysis result of the graphical programming job with the analyzable format. The technical problem that the teaching effect and the learning experience of students are affected due to inconsistent scores caused by subjective factors is solved.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence, and in particular to a method and device for correcting graphical programming assignments based on artificial intelligence. Background Art

[0002] With the rapid development of information technology, programming education has gradually become an important way to cultivate children's logical thinking and innovation ability. Especially in children's programming education, graphical programming, as an intuitive and easy-to-understand method, has occupied an important position in the field of education. Graphical programming greatly reduces the threshold for programming learning by dragging and splicing code blocks, allowing children to be exposed to programming concepts at an early age and gradually master programming skills in practice. This method can not only stimulate children's interest, but also help them better understand abstract programming logic. Therefore, it has been widely used in children's programming education.

[0003] However, with the popularity of programming education and the surge in the number of participants, teachers are facing unprecedented challenges in grading graphical programming homework. The traditional manual grading method cannot meet the needs of grading large-scale student homework. Teachers need to check each homework one by one, which takes a lot of time and energy. Especially in the case of a large class size, the teacher's workload is increased, and it is difficult to ensure that every student can get timely and accurate feedback. This manual grading method is not only inefficient, but may also lead to inconsistent grading due to subjective differences among teachers, affecting the fairness and accuracy of grading. At the same time, it is difficult for teachers to provide personalized guidance and feedback to each student in a timely manner, limiting students' opportunities to quickly correct errors and improve their programming skills.

[0004] The current teaching model still has several significant shortcomings: First, the traditional manual grading method is inefficient and cannot cope with the needs of grading large-scale student assignments. Teachers need to spend a lot of time checking the assignments one by one, which reduces teaching efficiency. Secondly, due to the lack of unified grading standards and norms, different teachers may have different evaluations of the same assignment, which affects the fairness and accuracy of the grading. Thirdly, manual grading usually takes a long time, and students cannot get feedback in time, which not only affects students' learning progress, but also limits their opportunities to quickly correct mistakes. Finally, due to limited educational resources, teachers cannot provide high-quality, personalized guidance for each student, resulting in uneven resource allocation and failure to meet the learning needs of each student. These problems have affected the effectiveness of graphical programming education and students' learning experience to a certain extent.

[0005] Therefore, how to improve the efficiency, consistency and timeliness of graphical programming homework grading while ensuring the personalization and high quality of feedback has become a pressing issue to be addressed in the current education field.

[0006] To address the above-mentioned problems, no effective solution has been proposed yet. Summary of the invention

[0007] The embodiments of the present invention provide a graphical programming homework grading method and device based on artificial intelligence, so as to at least solve the technical problem that subjective factors lead to inconsistent grading and affect the teaching effect and students' learning experience.

[0008] According to one aspect of an embodiment of the present invention, there is provided a method for correcting graphical programming jobs based on artificial intelligence, including: obtaining a graphical programming job, and converting the graphical programming job into a graphical programming job in an analyzable format; based on a pre-built knowledge base, using the AI ​​big model to perform correction analysis on the graphical programming job in the analyzable format, and obtaining a correction analysis result of the graphical programming job in the analyzable format.

[0009] According to another aspect of an embodiment of the present invention, there is also provided a graphical programming platform based on artificial intelligence, including: an acquisition module, configured to acquire a graphical programming job and convert the graphical programming job into a graphical programming job in an analyzable format; a correction module, configured to perform correction and analysis on the graphical programming job in the analyzable format using the AI ​​big model based on a pre-built knowledge base, and obtain a correction and analysis result of the graphical programming job in the analyzable format.

[0010] In an embodiment of the present invention, a graphical programming assignment is obtained and converted into a graphical programming assignment in an analyzable format; based on a pre-built knowledge base, the AI ​​big model is used to perform correction analysis on the graphical programming assignment in the analyzable format, and the correction analysis result of the graphical programming assignment in the analyzable format is obtained. Through the above scheme, the technical problem of inconsistent scoring caused by subjective factors, which affects the teaching effect and students' learning experience, is solved. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of this application. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0012] Figure 1 is a flow chart of a graphical programming homework grading method based on artificial intelligence according to an embodiment of the present invention;

[0013] Figure 2 is an architecture diagram of an artificial intelligence-based graphical programming platform according to an embodiment of the present invention;

[0014] Figure 3is a flow chart of another method for correcting graphical programming homework based on artificial intelligence according to an embodiment of the present invention;

[0015] Figure 4 is an architecture diagram of another graphical programming platform based on artificial intelligence according to an embodiment of the present invention;

[0016] Figure 5 A schematic diagram of the structure of an electronic device suitable for implementing the embodiments of the present disclosure is shown. DETAILED DESCRIPTION

[0017] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0018] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0019] According to an embodiment of the present invention, a method embodiment of a graphical programming homework grading method based on artificial intelligence is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0020] Figure 1 A graphical programming homework correction method based on artificial intelligence according to an embodiment of the present invention is as follows: Figure 1 As shown, the method comprises the following steps:

[0021] Step S102, obtaining a graphical programming job, and converting the graphical programming job into a graphical programming job in an analyzable format.

[0022] After completing the graphical programming assignment, students upload the assignment to the system through the platform interface. Assignments can be obtained by directly importing files or interacting with other programming platforms through the platform's API interface. The uploaded assignment information will be stored in the system for subsequent processing and use.

[0023] After the job is successfully uploaded, the job processing module converts the graphical programming job into a format that can be analyzed. Graphical programming jobs are usually composed of different building blocks, each of which represents an operation or logical structure. Since these building blocks are represented in a graphical way and cannot be directly understood by the computer system, they need to be converted into a structured data format that can be processed by the machine. Specifically, first, the system will parse the uploaded graphical programming job and identify each building block in it. After the parsing is completed, the job processing module will convert the graphical building blocks into a structured data format according to preset rules and standards. Common formats include graphical code sequences (such as JSON, XML format), or abstract syntax tree (AST) forms, which represent the execution flow of the program, the connection relationship between building blocks and their parameters. During the conversion process, the job processing module will also standardize and optimize the data to ensure that the job data of different graphical programming platforms can be processed uniformly. Standardization includes unified data format, unified naming specifications of building block types, etc., so that subsequent models can be analyzed more efficiently and accurately.

[0024] Step S104, based on the pre-built knowledge base, using the AI ​​big model to perform correction analysis on the graphical programming job in the analyzable format, to obtain the correction analysis result of the graphical programming job in the analyzable format.

[0025] First, the knowledge base is constructed based on the content of the graphical programming course, wherein the knowledge base includes at least one of the following: programming concepts, logical structures, error types and solutions. The knowledge base of the programming concept part includes functional descriptions of different types of building blocks, such as variable blocks, operation blocks, conditional judgment blocks, loop blocks, etc., and how these building blocks are used in combination. The syntax, usage, parameter configuration and role of each building block in the program will be described in detail in the knowledge base to ensure that the large model can accurately identify these basic components. The logical structure part is used to describe common control structures in programming languages, such as sequential execution, conditional judgment and loop structures. In graphical programming, logical structures are usually implemented by different types of building blocks. The knowledge base defines the application rules and specifications of these control structures in detail to help the large model determine whether these structures are used correctly in students' homework. For example, whether the loop blocks are connected correctly, whether the conditional judgment blocks are configured with reasonable judgment conditions, etc. In addition, the knowledge base will also cover complex programming logic such as nested structures and recursive calls to ensure that the large model can accurately analyze the program flow in complex homework. Error types are also a key component of the knowledge base. In graphical programming, students are prone to various common errors, such as incorrect connection of building blocks, incorrect data flow, irregular loop structure, etc. The knowledge base lists these common error types and provides detailed solutions for each error. For example, if a student fails to correctly connect the conditional judgment building blocks in an assignment, the knowledge base will provide suggestions for solving this problem, such as how to rearrange the building blocks, adjust parameter settings, etc. In addition, the knowledge base will also include some typical error repair methods, common programming traps and optimization suggestions to help large models make more accurate judgments when performing error detection.

[0026] Next, an AI big model is constructed and trained. The AI ​​big model is constructed based on the transformer network; the AI ​​big model is trained so that the AI ​​big model can understand the logical structure of the graphical programming language and can automatically identify the functions and errors of the graphical programming program.

[0027] Then, the trained AI big model is used for correction analysis. For example, based on the job content of the graphical programming job in the analyzable format, the knowledge base is queried in real time to obtain corresponding knowledge, wherein the corresponding knowledge includes knowledge points and / or cases; based on the corresponding knowledge, the AI ​​big model is used to perform correction analysis on the graphical programming job in the analyzable format in at least one of the following ways: parsing code logic, detecting errors, and evaluating job quality.

[0028] Afterwards, based on the correction analysis results, a personalized correction report and learning suggestions are generated; a review of the personalized correction report and learning suggestions is received, and the personalized correction report and learning suggestions after the review are fed back. The personalized correction report and learning suggestions include at least one of the following: error location, modification suggestions, and learning resource links.

[0029] The present invention provides a graphical programming homework grading platform based on a large model, which optimizes the grading process through artificial intelligence technology and improves the efficiency, accuracy and personalization of grading. Figure 2 As shown, the platform architecture of the present invention is divided into five main layers, namely user interface layer 20, job processing layer 22, large model analysis layer 24, knowledge base service layer 26 and report generation and feedback layer 28. Each layer implements different functions to ensure that the platform can efficiently and accurately correct graphical programming homework and provide personalized feedback for students and teachers. The platform provided in this embodiment can automatically correct graphical programming homework based on large-scale pre-trained language models (such as Transformer, etc.).

[0030] The user interface layer 20 mainly includes a student homework upload interface and a teacher correction result viewing interface. Students can submit their graphical programming homework through this interface, and support importing homework files from multiple programming platforms. Teachers can view the correction results and related statistical data through this interface, review AI correction opinions, and add personal comments or modification suggestions.

[0031] The assignment processing layer 22 is used to convert the graphical programming assignments submitted by students into a data format that can be analyzed by the large model. Graphical programming assignments use blocks to represent code logic, and the assignment processing module converts these blocks into structured data (such as tree diagrams, graphical code sequences, or intermediate representation codes) for further processing by the large model. The converted data will be sent to the large model evaluation module for analysis.

[0032] The big model analysis layer 24 is the core of the platform and is responsible for in-depth analysis of students' assignments. This module integrates automated analysis algorithms based on big models (such as Transformer), which can understand the logical structure of graphical programming languages ​​and identify program functions, error types, and potential optimization spaces in assignments. During the training process of the big model, a large number of programming examples and error type cases are used, enabling it to accurately identify common logical errors, syntax problems, and inefficient programming structures.

[0033] The knowledge base service layer 26 provides resources such as basic knowledge, error types, and best practices related to programming. The knowledge base of this layer contains common programming concepts, programming errors and their solutions, optimization suggestions, teaching resources, etc., which supports real-time query of large models when performing job analysis and assists the model in making more accurate judgments. For example, when a certain building block combination in a job has a common error pattern, the knowledge base can provide relevant solutions or correction methods to help the model better analyze and correct the error.

[0034] The report generation and feedback layer 28 automatically generates a detailed correction report based on the analysis results of the large model. The report content includes the location of errors in the homework, modification suggestions, learning resource links, etc. The generated report will provide personalized learning suggestions based on the characteristics of the student's homework, helping students understand the errors and improve their programming skills. At the same time, the feedback generation module can also generate reports with pictures and texts to enhance the student's learning experience.

[0035] The following will describe the workflow of the graphical programming platform in detail. Figure 3 As shown, the method for correcting graphical programming homework based on artificial intelligence applied on the graphical programming platform includes the following steps:

[0036] Step S302, receiving homework submitted by students.

[0037] After completing the graphical programming assignment, students submit the assignment through the platform interface. The platform supports uploading assignment files of various graphical programming platforms. The assignment content can be obtained by directly importing the file or interacting with the platform's API interface. The uploaded assignment information will be stored in the platform for subsequent processing.

[0038] Step S304, converting the job format.

[0039] After students submit their completed graphical programming assignments through the platform interface, the assignment processing module receives the uploaded assignment files. Since graphical programming languages ​​use building blocks to represent code and logical structures, the task of the assignment processing module is to convert these graphical programming contents into a structured data format that can be understood and analyzed by computers. Specifically, the following steps are included:

[0040] 1) Job file analysis.

[0041] Graphical programming jobs are usually composed of different building blocks, each of which represents a specific operation or logical relationship (such as variable definition, loop structure, conditional judgment, etc.). The job processing module first parses the uploaded job file and identifies each building block in the job. The building block format of different graphical programming platforms may be different, but the job processing module parses according to the interface or file structure provided by the platform to ensure that all necessary job content can be accurately extracted.

[0042] Specifically, the job processing module uses preset rules and standardized methods to parse each building block in the graphical programming into a unified format. The preset rules include building block type classification, data type conversion, function call identification, etc. Through the above standardized processing, the platform can be compatible with jobs of different graphical programming platforms, ensuring that jobs generated by graphical programming tools can be correctly parsed and processed.

[0043] 2) Generate intermediate representation.

[0044] In order to enable the job data to be processed by the big model, the job processing module converts these graphical building blocks into a standardized intermediate representation. The intermediate representation includes a graphical code sequence (such as XML format, JSON format) or a tree diagram (such as an abstract syntax tree). The structured data form can clearly express the function, operation sequence and relationship of each building block, ensuring that the big model can effectively analyze and evaluate it.

[0045] In some embodiments, the job processing module can also convert each building block into a corresponding code snippet, and the snippet can use standardized syntax (such as XML or JSON format) to represent the type, parameters, connection method and other information of the building block. In another embodiment, a structured representation method can also be used to convert the building blocks in the job into a tree structure, where each node represents a building block or a logical unit, and the edge represents the connection relationship between the building blocks. The tree diagram can clearly show the execution flow of the code and help the analysis module identify the logical structure of the code.

[0046] Through the above steps, the job processing module effectively converts the graphical programming jobs from intuitive building blocks to structured data suitable for machine analysis. This conversion not only improves the accuracy of job analysis, but also lays a solid foundation for subsequent automatic correction, report generation and personalized feedback.

[0047] Step S306, large model evaluation.

[0048] The converted job data is transmitted to the big model. During the analysis process, the big model will combine its own learned capabilities and the programming knowledge provided in the knowledge base to deeply analyze the code logic in the job data and perform error detection, function identification, and creativity and complexity assessment.

[0049] In some embodiments, the core function of the large model comes from its pre-training ability. By learning from a large number of programming cases, it can identify common problems in graphical programming languages, such as incorrect building block connections, incorrect data flow direction, and irregular loop structures. During the job evaluation process, the model relies on its built-in knowledge to understand the overall structure and function of the code, evaluate whether the program in the job conforms to the programming logic and specifications, and identify possible errors. Specifically, the first eigenvector of the context code of the graphical programming job and the second eigenvector of the current code to be detected are extracted respectively, and the first eigenvector and the second eigenvector are respectively subjected to the order-up processing according to the extended rank criterion, and the multi-channel convolution kernel is used to perform feature coupling and fusion processing of the first eigenvector and the second eigenvector to obtain the coupled main embedded feature matrix and the auxiliary embedded feature matrix. The above matrix is ​​expanded along the direction of the extended dimension to form multiple sub-matrices. Each sub-matrix is ​​subjected to singular value decomposition to extract the core subspace representation and factor matrix. The main component difference between the main feature matrix and the auxiliary feature matrix is ​​calculated based on the core subspace representation and the factor matrix to extract the core features, thereby effectively identifying logical errors and reference problems.

[0050] In some other embodiments, the big model can also be evaluated without relying on the data in the pre-training phase. During the evaluation process, the model uses the relevant knowledge in the knowledge base as input parameters to help the big model make more accurate judgments. The knowledge base stores detailed rules of programming languages, common errors and their correction solutions, implementation details of programming logic, etc. This knowledge plays an auxiliary role in the homework grading process. The big model combines this knowledge and can identify logical errors, redundant operations, structures that do not meet programming specifications, and other problems in the code through in-depth analysis of student homework.

[0051] For example, if a student uses an improper loop structure or incorrectly connects building blocks in an assignment, the knowledge base will provide relevant programming rules and correction methods, and the big model will judge the error based on these rules. By combining these external inputs from the knowledge base, the big model can identify problems in the assignment and generate correction suggestions.

[0052] Specifically, when the big model evaluates homework, it will analyze the types of errors in the program based on the knowledge provided by the knowledge base. For example, if the model identifies a problem with the loop structure, it will obtain rules about "how to build a correct loop structure" from the knowledge base and feed this information back into the evaluation process. In this way, the model can more accurately determine the cause of the error and provide targeted modification suggestions. At the same time, the knowledge base may also contain relevant programming examples or tutorials to help students further understand their errors and correct them.

[0053] The following is an example of knowledge in the knowledge base:

[0054] / *When triggering an event: This event is triggered when the specified event type (click, press, release) occurs on the specified target (self, screen, a character)

[0055] *parameter:

[0056] *eventType: event type, which can be "click", "press" or "release".

[0057] *target: event target, which can be "self", "screen" or a specific role.

[0058] * /

[0059] typeWhen triggering event=(

[0060] eventType: "click" | "press" | "release",

[0061] target:"self"|"screen"|string

[0062] )=>void

[0063] Finally, after the big model completes the homework analysis, the feedback generation module will automatically generate a detailed correction report based on the evaluation results of the big model. The report includes error location, correction suggestions, and links to learning resources. The report not only lists the specific errors made by students in their homework, but also provides further learning resources, such as programming tutorials, solutions, and sample codes, to help students understand the causes of the errors and improve them. In this way, students can learn more programming knowledge from their mistakes and gradually improve their programming skills.

[0064] By combining the powerful capabilities of the big model and the rich resources of the knowledge base, the homework grading system of the present invention can efficiently and accurately evaluate homework and provide students with personalized feedback and improvement suggestions. The grading method based on the big model and knowledge base provided by the present invention not only improves the grading efficiency, but also provides students with more targeted learning support, which helps to improve the quality of programming education and learning effects.

[0065] Step S308: teacher review and interaction.

[0066] The generated correction report will be presented to teachers through the teacher review and interaction module. Teachers can view the AI ​​automatic correction results by logging into the system and conduct a comprehensive review. The report generated by the platform contains the analysis and evaluation results of the student's homework by the big model. Teachers can choose to adjust or supplement these correction results according to the actual situation. Teachers can not only view the errors identified in the homework, but also provide personalized feedback and guidance based on their understanding of the students to ensure the accuracy and pertinence of the correction content.

[0067] During the teacher review process, teachers can add personalized comments to the report based on the specific circumstances of the student's assignment. For example, if the teacher believes that the AI ​​analysis does not fully cover certain details in the assignment or misses some errors, the teacher can modify, supplement or adjust it directly in the report. Teachers can also give additional learning suggestions for students' performance to help them further improve their programming skills, or provide individual guidance for students' specific problems.

[0068] In addition, teachers can also provide detailed supplementary explanations in the report for students' special problems. For example, if students have obvious difficulties with a certain programming concept, teachers can explain the cause of the problem in detail in the report and guide students on how to fill this knowledge gap through additional learning resources. Teachers can recommend relevant teaching materials, tutorials or programming examples to students through the system to further help students consolidate what they have learned.

[0069] After the teacher completes the review and revision, the platform will update the correction report and pass the final feedback to the students through the interactive interface between teachers and students. The interaction between teachers and students is not limited to simple revisions or supplements. Teachers can also communicate with students more directly through online comments, instant messages, etc. to ensure that students can understand the feedback and make targeted learning improvements based on the feedback.

[0070] Through this interactive mechanism, teachers can not only ensure the accuracy and quality of AI grading results, but also provide personalized guidance and suggestions based on students' specific needs. This application combines teachers with AI grading results, which not only improves the accuracy and efficiency of grading, but also provides students with more personalized and comprehensive learning support, further promoting the personalized development of programming education.

[0071] This application also provides a graphical programming platform based on artificial intelligence, such as Figure 4 As shown, it includes: an acquisition module 42, which is configured to acquire a graphical programming job and convert the graphical programming job into a graphical programming job in an analyzable format; a correction module 44, which is configured to perform correction analysis on the graphical programming job in the analyzable format based on a pre-built knowledge base using the AI ​​big model to obtain a correction analysis result of the graphical programming job in the analyzable format.

[0072] It should be noted that: the graphical programming platform based on artificial intelligence provided in the above embodiment is only illustrated by the division of the above functional modules. In actual application, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. In addition, the graphical programming platform based on artificial intelligence provided in the above embodiment and the graphical programming method embodiment based on artificial intelligence belong to the same concept. The specific implementation process is detailed in the method embodiment and will not be repeated here.

[0073] Figure 5 FIG. 1 shows a schematic diagram of the structure of an electronic device suitable for implementing the embodiment of the present disclosure. It should be noted that: Figure 5 The electronic device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present disclosure.

[0074] like Figure 5 As shown, the electronic device includes a central processing unit (CPU) 1001, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 1002 or a program loaded from a storage part 1008 into a random access memory (RAM) 1003. In the RAM 1003, various programs and data required for system operation are also stored. The CPU 1001, the ROM 1002, and the RAM 1003 are connected to each other via a bus 1004. An input / output (I / O) interface 1005 is also connected to the bus 1004.

[0075] The following components are connected to the I / O interface 1005: an input section 1006 including a keyboard, a mouse, etc.; an output section 1007 including a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker, etc.; a storage section 1008 including a hard disk, etc.; and a communication section 1009 including a network interface card such as a LAN card, a modem, etc. The communication section 1009 performs communication processing via a network such as the Internet. A drive 1010 is also connected to the I / O interface 1005 as needed. A removable medium 1011, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 1010 as needed, so that a computer program read therefrom is installed into the storage section 1008 as needed.

[0076] The above is only a preferred implementation of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present application. These improvements and modifications should also be regarded as the scope of protection of the present application.

Claims

1. A graphical programming homework grading method based on artificial intelligence, characterized in that: include: Acquire a graphical programming job, and convert the graphical programming job into a graphical programming job in a parseable format; Based on the pre-built knowledge base, the AI ​​big model is used to perform correction analysis on the graphical programming job in the analyzable format to obtain the correction analysis result of the graphical programming job in the analyzable format.

2. The method according to claim 1, characterized in that Based on the pre-built knowledge base, the AI ​​big model is used to perform correction and analysis on the graphical programming homework in the analyzable format, including: Based on the job content of the graphical programming job in the analyzable format, query the knowledge base in real time to obtain corresponding knowledge, wherein the corresponding knowledge includes knowledge points and / or cases; Based on the corresponding knowledge, the AI ​​big model is used to perform at least one of the following correction and analysis on the graphical programming job in the analyzable format: parsing code logic, detecting errors, and evaluating job quality.

3. The method according to claim 1, characterized in that After obtaining the correction analysis result of the graphical programming job in the analyzable format, the method further includes: Based on the correction analysis results, generate personalized correction reports and learning suggestions; Receive a review of the personalized correction report and learning suggestions, and provide feedback on the reviewed personalized correction report and learning suggestions.

4. The method according to claim 1, characterized in that: Before using the AI ​​big model to correct and analyze the graphical programming homework in the analyzable format, the method also includes: building the knowledge base based on the content of the graphical programming course, wherein the knowledge base includes at least one of the following: programming concepts, logical structures, error types and solutions.

5. The method according to claim 1, characterized in that Before using the AI ​​big model to perform correction analysis on the graphical programming homework in the analyzable format, the method further includes: Building the AI ​​big model based on the transformer network; The AI ​​big model is trained so that the AI ​​big model can understand the logical structure of the graphical programming language and can automatically identify the functions and errors of the graphical programming program.

6. The method according to claim 3, characterized in that The personalized correction report and learning suggestions include at least one of the following: error location, modification suggestions and learning resource links.

7. A graphical programming platform based on artificial intelligence, characterized in that: include: An acquisition module is configured to acquire a graphical programming job and convert the graphical programming job into a graphical programming job in a parseable format; The correction module is configured to perform correction analysis on the graphical programming job in the analyzable format using the AI ​​big model based on a pre-built knowledge base, and obtain the correction analysis result of the graphical programming job in the analyzable format.

8. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored program, wherein when the program is executed, the device where the computer-readable storage medium is located is controlled to execute the method according to any one of claims 1 to 6.

9. A computer device, characterized in that: include: Memory and processor, The memory stores a computer program; The processor is used to execute the computer program stored in the memory, and when the computer program is run, the processor is enabled to execute the method according to any one of claims 1 to 6.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images