Intelligent construction and interactive display system for course knowledge graph

Abstract

The invention discloses an intelligent construction and interactive display system for a course knowledge graph, and aims to design a course ontology to ensure systematicness and comprehensiveness of course knowledge; the invention provides an intelligent course knowledge extraction method based on a large model, and aims to intelligently extract knowledge key points in courses so as to improve the knowledge extraction efficiency and accuracy. A multi-source database storage mode and convenient data access operation are provided, the problem that a traditional graph database is single in storage content and inconvenient for a client to operate data is solved, and efficient storage and management of the graph database and a relational database in a multi-source data scene are achieved. According to the system, course knowledge points can be efficiently and accurately extracted, dynamic fusion and real-time updating of multi-source heterogeneous data are supported, and the user experience is improved through an interactive display mode.

Description

Technical Field

[0001] The present invention relates to the field of intelligent extraction, storage, and display methods and system technologies for course knowledge graphs, and in particular to an intelligent construction and interactive display system for course knowledge graphs. Background Art

[0002] With the development of big data and artificial intelligence technology, knowledge graphs have been widely used as a new type of knowledge management and data expression. Especially in the field of education, by structured expression of course knowledge points and their relationships, knowledge graphs can organize scattered course information into an orderly and highly related knowledge network. This structured expression method can not only help teachers and students clearly understand the connection between knowledge points, but also assist the system to make intelligent recommendations based on students' knowledge mastery and optimize the arrangement of teaching content. In addition, knowledge graph technology can also support personalized learning path planning, dynamically adjust course content and learning order according to the needs and ability levels of different learners, and improve learning effects.

[0003] However, the construction, storage and display of current course knowledge graphs still face the following challenges:

[0004] 1. In terms of course ontology construction, there is a huge and complex knowledge system between different courses and disciplines. How to effectively classify and organize this knowledge and construct a reasonable knowledge system is a key challenge facing existing education platforms;

[0005] 2. In terms of intelligent extraction of knowledge points, existing extraction methods mostly rely on natural language processing technology and rule matching algorithms. However, in the face of semantically complex and diverse course content, existing technologies still have limitations in accuracy and efficiency, especially the inability to accurately capture knowledge points implicit in the context;

[0006] 3. In terms of data storage, most existing storage methods are based on relational databases or simple document storage, which cannot efficiently support the integration and real-time update of multi-source data, resulting in data redundancy and inconsistency problems;

[0007] 4. In terms of graph display and interaction, the current knowledge graph display methods are mostly static, lacking interaction with users, and unable to dynamically adjust the display content according to the user's specific learning needs. In addition, the existing display methods have limited visualization effects on complex relationships, affecting users' comprehensive understanding of the course knowledge system;

[0008] Therefore, we designed an intelligent construction and interactive display system for course knowledge graphs to solve the above problems. Summary of the invention

[0009] The purpose of the present invention is to solve the shortcomings of the prior art and to propose an intelligent construction and interactive display system for a course knowledge graph. The system of the present invention can efficiently and accurately extract course knowledge points, support dynamic fusion and real-time updating of multi-source heterogeneous data, and improve user experience through interactive display.

[0010] In order to achieve the above object, the present invention adopts the following technical solutions:

[0011] An intelligent construction and interactive display system for course knowledge graphs, including the following purposes:

[0012] Design the course ontology to ensure the systematic and comprehensiveness of course knowledge;

[0013] This paper provides an intelligent course knowledge extraction method based on a large model, which aims to intelligently extract the key points of knowledge in the course, thereby improving the efficiency and accuracy of knowledge extraction;

[0014] Provides multi-source database storage and convenient data access operations, solving the problem that traditional graph databases store single content and are inconvenient for clients to operate data, and realizing efficient storage and management of graph databases and relational databases in multi-source data scenarios;

[0015] Design a knowledge graph interactive system that can be dynamically displayed and operated, which can better display the nodes and relationships stored in the knowledge graph, edit and delete the nodes and relationships, and improve the visualization effect and operation experience.

[0016] Preferably, the course ontology combines the course type, teaching material catalog and internal logic, and is divided into five entities: course type, course, knowledge system, knowledge unit and knowledge key points according to different knowledge granularity;

[0017] Course type: Course type refers to the type of course, which is intended to distinguish and standardize different types of courses, such as basic courses, advanced courses, and special courses. Course type helps to clarify the positioning of courses in teaching activities, and by establishing associations with specific course entities, ensures that the classification of each course is clear and the attributes are standardized;

[0018] The course: a specific course entity, including the course name, course introduction, reference materials, course objectives, learning difficulty, course credits, assessment method, pilot courses and related learning resources. Course entities are linked through attributes (such as "pilot courses") and correspond to "course types", which usually correspond to specific course content in textbooks, and provide detailed information on credits, lecturers and related resources to ensure the unity of learning objectives and assessment standards;

[0019] The knowledge system is used to define the level of the abstract knowledge system in the course, including the name of the knowledge system, the importance, the next knowledge system, and the attributes of the course to which it belongs. The knowledge systems are connected in order through the "next knowledge system" attribute, showing the progressive level and logical structure of knowledge. Through such division, the knowledge system clarifies the organization of the core knowledge in each course, making its content more systematic;

[0020] The knowledge unit: the knowledge unit is a further refinement of the knowledge system, usually corresponding to the program directory in the textbook. Each knowledge unit entity includes the knowledge unit name, knowledge unit content, importance, learning difficulty, and the attributes of the knowledge system to which it belongs. The knowledge unit entity is hierarchically managed through the knowledge system to which it belongs;

[0021] The multiple knowledge points are the smallest knowledge granularity in the course ontology. Each knowledge point represents a specific knowledge point and skill. The knowledge point entity includes the name of the knowledge point, the content of the knowledge point, the importance, the learning difficulty, the learning objectives, and the attributes of the knowledge unit to which it belongs. The knowledge point entity is associated with the parent knowledge unit through the attributes of the knowledge unit to which it belongs.

[0022] Preferably, in order to realize the hierarchical organization and automated management of knowledge, the system introduces six key relationship types, namely, the type of course, the predecessor course, the course to which it belongs, the next knowledge system, the knowledge system to which it belongs, and the knowledge unit to which it belongs. The following describes these relationships in detail and their applications at each level;

[0023] The course type is used to connect the course and its type to identify the specific category to which each course belongs; through this relationship, the system can effectively classify and manage courses;

[0024] The said pre-courses: reveal the learning sequence between courses and clarify the prerequisite courses that need to be mastered before learning a certain course. Through this relationship, the system can ensure the rationality of the learning path and prevent students from entering difficult courses without the necessary basic knowledge;

[0025] The courses to which it belongs are used to associate a knowledge system with its courses. Each course contains one or more knowledge systems, and a knowledge system represents the core content and knowledge framework of the course.

[0026] The next knowledge system: defines the order between knowledge systems, ensuring that students can follow the correct logic or difficulty order when learning a certain knowledge system. Through this relationship, the system can plan a coherent learning path for students to avoid knowledge gaps;

[0027] The knowledge system to which it belongs is used to connect the knowledge unit and the knowledge system to which it belongs, so as to identify which knowledge system the specific knowledge unit belongs to. The knowledge unit is a further subdivision of the knowledge system and covers more specific content.

[0028] The belonging knowledge unit: connect the knowledge points with the corresponding knowledge units to clarify which knowledge unit the specific knowledge points belong to. As the smallest unit of the knowledge graph, the knowledge points usually represent specific concepts or skills.

[0029] Preferably, the intelligent course knowledge extraction method based on a large model is specifically divided into three steps: knowledge division and precise extraction, large model-assisted attribute identification and knowledge summary, and knowledge graph relationship construction and hierarchical organization;

[0030] Knowledge division and accurate extraction of course content based on regular expressions: Based on ontology design and rule-driven, regular expressions are used to extract courses, knowledge systems, knowledge units and key points from texts, providing basic data support for building a complete knowledge graph and ensuring that the knowledge graph structure is clear and orderly;

[0031] Big model assists entity attribute recognition and knowledge key point summary: In the specific implementation process of the present invention, regular expressions first realize the preliminary extraction of course content, but for complex knowledge key points and attributes, the accuracy of relying solely on rule matching is limited. Therefore, the system introduces a big model and uses natural language processing technology to further identify entity attributes and summarize knowledge key points;

[0032] Knowledge graph relationship construction and hierarchical organization: Using the course data obtained in the previous step, a unified knowledge system name is set as the basic structure, and three entities, namely knowledge system, knowledge unit and knowledge key points, are structured and organized. Under the ontology design, specific attributes are preset for each entity, including sequential numbering, to achieve modularization and systematic processing of knowledge content. Finally, the course content is decomposed into standardized data files level by level, laying the foundation for the hierarchical organization and visualization construction of the knowledge graph;

[0033] Through the above steps, the course content is decomposed, extracted, and organized into standardized data files, laying the foundation for the visual construction of the course knowledge graph.

[0034] Preferably, in order to achieve the third purpose of the present invention, the Neo4j graph database is used for storage, py2neo is used to perform programmatic graph data operations, and finally the FlaskAPI is used to provide an interface for external calls. The graph data storage provided by Neo4j is particularly suitable for the storage needs of knowledge graphs. The FlaskAPI can cooperate with the py2neo library to conveniently operate Neo4j in a programmatic manner and provide a REST API interface to the outside world. Finally, the user identity and authentication information are stored in the relational database MySQL and a high-performance interface is provided through SpringBoot.

[0035] Preferably, in order to achieve the fourth object of the present invention, the following steps are included:

[0036] S1, obtain the course nodes and relationship content of the knowledge graph, use Echarts to render different node types accordingly, and improve the visualization effect of the knowledge graph and the recognition of the node type hierarchy;

[0037] S2, in the graph display component, register the click events of each node and relationship line, dynamically switch to the next level of nodes and relationships of the clicked element by clicking or dragging the nodes and relationship lines, and obtain the node and relationship data saved in the graph database in the backend FlaskAPI, and dynamically render the returned results to the knowledge graph display area, making the interactive experience simple and intuitive;

[0038] S3, in the graph editing component, register the click events of nodes and relationship lines to realize the click selection of nodes or relationships to be edited or deleted. After the operation, it will take effect immediately and dynamically on the graph editing component. The operation is convenient and the feedback is intuitive.

[0039] Compared with the prior art, the present invention has the following beneficial effects:

[0040] 1. Multi-level knowledge graph construction: By combining regular expressions and large models, knowledge points can be extracted efficiently and accurately from course materials and a multi-level knowledge graph can be constructed.

[0041] 2. Intelligent extraction and attribute recognition: Use large models (such as ChatGPT4) to intelligently extract key knowledge points and recognize attributes to improve the automation and accuracy of knowledge graph construction.

[0042] 3. Intelligent extraction and attribute recognition: Use large models (such as ChatGPT4) to intelligently extract key knowledge points and recognize attributes to improve the automation and accuracy of knowledge graph construction.

[0043] 4. Dynamic display and interactive system: A knowledge graph interactive system with dynamic display and operation is designed, using Echarts to enhance visualization effects, allowing users to easily edit and manage knowledge points.

[0044] In summary, the system of the present invention can extract course knowledge points efficiently and accurately, support dynamic fusion and real-time updating of multi-source heterogeneous data, and improve user experience through interactive display. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 This is a structural diagram of an intelligent construction and interactive display system of a course knowledge graph proposed by the present invention;

[0046] Figure 2 This is a structural schematic diagram of the transmission mechanism in the intelligent construction and interactive display system of the course knowledge graph proposed by the present invention. DETAILED DESCRIPTION

[0047] The technical solutions in the embodiments of the present invention will be described clearly and completely 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, rather than all the embodiments.

[0048] Reference Figure 1-Figure 2 , an intelligent construction and interactive display system for course knowledge graphs, including the following purposes:

[0049] Design the course ontology to ensure the systematic and comprehensiveness of course knowledge;

[0050] This paper provides an intelligent course knowledge extraction method based on a large model, which aims to intelligently extract the key points of knowledge in the course, thereby improving the efficiency and accuracy of knowledge extraction;

[0051] Provides multi-source database storage and convenient data access operations, solving the problem that traditional graph databases store single content and are inconvenient for clients to operate data, and realizing efficient storage and management of graph databases and relational databases in multi-source data scenarios;

[0052] Design a knowledge graph interactive system that can be dynamically displayed and operated, which can better display the nodes and relationships stored in the knowledge graph, edit and delete the nodes and relationships, and improve the visualization effect and operation experience.

[0053] The course ontology combines the course type, textbook catalog and internal logic, and is divided into five entities: course type, course, knowledge system, knowledge unit and knowledge key points according to the different knowledge granularity;

[0054] Course type: Course type refers to the type of course, which aims to distinguish and standardize different types of courses, such as basic courses, advanced courses, and special courses. Course type helps to clarify the positioning of courses in teaching activities, and by establishing associations with specific course entities, ensure that the classification of each course is clear and the attributes are standardized;

[0055] Course: Specific course entity, including course name, course introduction, reference materials, course objectives, learning difficulty, course credits, assessment method, pilot courses and related learning resources. Course entities are linked through attributes (such as "pilot course") and correspond to "course type", which usually corresponds to the specific course content in the textbook, and provides detailed information on credits, lecturers and related resources to ensure the unity of learning objectives and assessment standards;

[0056] Knowledge system: used to define the level of abstract knowledge system in the course, including the name of the knowledge system, importance, next knowledge system, and the attribute of the course to which it belongs. The knowledge systems are connected in order through the attribute of "next knowledge system", showing the progressive level and logical structure of knowledge. Through such division, the knowledge system clarifies the organization of core knowledge in each course, making its content more systematic;

[0057] Knowledge unit: A knowledge unit is a further refinement of the knowledge system, usually corresponding to the program directory in the textbook. Each knowledge unit entity contains the knowledge unit name, knowledge unit content, importance, learning difficulty, and the attributes of the knowledge system to which it belongs. The knowledge unit entity is hierarchically managed through the knowledge system to which it belongs;

[0058] Multiple knowledge points: the smallest knowledge granularity in the course ontology. Each knowledge point represents a specific knowledge point and skill. The knowledge point entity includes the name of the knowledge point, the content of the knowledge point, the importance, the learning difficulty, the learning objectives, and the attributes of the knowledge unit to which it belongs. The knowledge point entity is associated with the parent knowledge unit through the attributes of the knowledge unit to which it belongs.

[0059] In order to realize the hierarchical organization and automatic management of knowledge, the system introduces six key relationship types, namely, the type of course, the predecessor course, the course to which it belongs, the next knowledge system, the knowledge system to which it belongs, and the knowledge unit to which it belongs. The following describes these relationships in detail and their applications at each level;

[0060] Course type: used to connect courses with their types to identify the specific category to which each course belongs; through this relationship, the system can effectively classify and manage courses. For example, one course can be identified as a "compulsory course for the major" and another course can be identified as an "elective course". This association not only helps learners quickly identify the importance of the course, but also provides support for course recommendation and screening;

[0061] Pre-course: reveals the order of learning between courses, and clarifies the prerequisite courses that need to be mastered before learning a certain course. Through this relationship, the system can ensure the rationality of the learning path and prevent students from entering more difficult courses without the necessary basic knowledge. For example, before learning "Linear Algebra", you may need to take "Advanced Mathematics". This relationship helps to design a reasonable course advancement path;

[0062] Courses: used to associate knowledge systems with their courses. Each course contains one or more knowledge systems, and a knowledge system represents the core content and knowledge framework of the course. Through this relationship, the system can modularize course content, help learners understand the course structure more clearly, and make course content management more efficient;

[0063] Next knowledge system: defines the order between knowledge systems, ensuring that students can follow the correct logic or difficulty order when learning a certain knowledge system. Through this relationship, the system can plan a coherent learning path for students to avoid knowledge gaps. For example, after learning "Matrix Basics", students should continue to learn "Matrix Operations". This relationship helps ensure the coherence of the learning process;

[0064] Knowledge system: used to connect a knowledge unit with the knowledge system it belongs to, to identify which knowledge system a specific knowledge unit belongs to. A knowledge unit is a further subdivision of a knowledge system, covering more specific content. Through this relationship, the system can refine the knowledge structure layer by layer, help learners deeply understand the subtle differences in knowledge points, and facilitate the management and display of knowledge units;

[0065] Belonging knowledge unit: Connect the key knowledge points with the corresponding knowledge units to clarify which knowledge unit the specific key knowledge points belong to. As the smallest unit of the knowledge graph, the key knowledge points usually represent specific concepts or skills. Through this relationship, the system can finely organize the smallest granularity of knowledge, ensure a clear knowledge structure, and help learners gradually master each knowledge point during retrieval and application.

[0066] A method for extracting knowledge from intelligent courses based on a big model is specifically divided into three steps: knowledge division and precise extraction, big model-assisted attribute identification and knowledge summarization, and knowledge graph relationship construction and hierarchical organization;

[0067] Knowledge division and accurate course content extraction based on regular expressions: Based on ontology design and rule-driven, regular expressions are used to extract courses, knowledge systems, knowledge units and key points from texts, providing basic data support for building a complete knowledge graph and ensuring that the knowledge graph structure is clear and orderly. The specific process is as follows:

[0068] Read electronic textbooks and convert them into standardized text strings to support subsequent semantic processing and modeling.

[0069] Input the name of the course's knowledge system (i.e., catalog information), and combine custom regular expressions and loop structures to accurately locate the start and end of each knowledge system. Based on the order of the knowledge system, segment the course text, and extract the name of each knowledge system and its detailed content.

[0070] The content of each knowledge system is processed, and the knowledge units are gradually extracted according to the hierarchical structure through regular expression technology. First, the main title (such as 1.1) is identified and extracted, followed by the extraction of subtitles (such as 1.1.1), and finally the detailed content of each knowledge point is extracted. The structure of each knowledge unit includes the name of the knowledge unit and its corresponding knowledge point. All extracted knowledge units and knowledge points are stored in the preset data structure according to the order of the corresponding knowledge system.

[0071] To ensure the accuracy and consistency of the order of the extracted results, the file contents are finally sorted. The sorting operation is performed according to the order of the knowledge system. By parsing the knowledge system number in the file name, the order of each knowledge system file is ensured to be consistent with the original structure of the course, thereby ensuring the logical consistency and hierarchy of the constructed knowledge graph.

[0072] Large model assists entity attribute recognition and knowledge key point summary: In the specific implementation process of the present invention, regular expressions first realize the preliminary extraction of course content, but for complex knowledge key points and attributes, the accuracy of simply relying on rule matching is limited. Therefore, the system introduces a large model and uses natural language processing technology to further identify entity attributes and summarize knowledge key points. The specific process is as follows:

[0073] The preprocessed hierarchical knowledge system content is first transferred to the big model, and the core attributes of each knowledge unit are identified through the big model. Specifically:

[0074] 1. Difficulty level: divided into three levels: low, medium and high;

[0075] 2. Importance: divided into three levels: low, medium and high;

[0076] 3. Main content of the knowledge unit: extract the core content related to the key points of knowledge.

[0077] Identify the core attributes of each knowledge point through the big model. Specifically:

[0078] 1. Difficulty level: divided into three levels: low, medium and high;

[0079] 2. Importance: divided into three levels: low, medium and high;

[0080] 3. Learning objectives: including theoretical learning, methodological learning, and applied practice;

[0081] 4. Main content of key knowledge points: Extract the core content related to the key knowledge points.

[0082] All identified knowledge units and knowledge key point attribute information will be stored in the dictionary of the corresponding entity, so that the entities and attributes are organized into a structured data format to facilitate subsequent knowledge graph construction and further processing.

[0083] Through this system design, complex knowledge points and attributes in the course content can be efficiently and accurately extracted by the large model, providing a solid foundation for the construction of knowledge graphs and teaching applications.

[0084] Knowledge graph relationship construction and hierarchical organization: Using the course data obtained in the previous step, a unified knowledge system name is set as the basic structure, and three entities, namely knowledge system, knowledge unit and knowledge key points, are structured and organized. Under the ontology design, specific attributes are preset for each entity, including sequential numbering, to achieve modularization and systematic processing of knowledge content. Finally, the course content is decomposed step by step into standardized data files, laying the foundation for the hierarchical organization and visual construction of the knowledge graph.

[0085] In the specific implementation process of the present invention, according to the knowledge graph ontology design, the course content is divided into three data templates, namely the knowledge system template, the knowledge unit template and the knowledge key point template.

[0086] In the knowledge system arrangement stage, the chapter information of the course is used as the basic unit of the knowledge system. Each knowledge system contains the name, learning order, importance, next module name, and the attribute of the course. The "learning order" number is automatically generated in the order of the chapters, and the sequential connection between knowledge systems is achieved through this attribute to form an orderly knowledge system chain.

[0087] In the knowledge unit sorting stage, each knowledge unit is associated with the knowledge system to which it belongs by reading the preprocessed knowledge system data. In the data structure, the name, difficulty level, importance level, knowledge point content, and course attributes of each knowledge unit are added. And the knowledge system to which the attributes in the knowledge unit belong is matched with the corresponding knowledge system to ensure the consistency of the knowledge unit and the knowledge system structure.

[0088] In the stage of knowledge key points arrangement, the knowledge content under each knowledge unit is further refined. By matching the preset data files, the knowledge key points information is matched with the knowledge units of the previous level one by one to ensure the subordinate relationship between the knowledge key points and the knowledge units. Each knowledge key point contains the name, content, knowledge unit to which it belongs, learning objectives, importance, difficulty and course attributes to fully describe the details of the knowledge key points.

[0089] Through the above steps, the course content is decomposed, extracted, and organized into standardized data files, laying the foundation for the visual construction of the course knowledge graph.

[0090] In order to achieve the third purpose of the present invention, Neo4j graph database is used for storage, py2neo is used to operate the graph data programmatically, and finally the FlaskAPI is used to provide an interface for external calls. The graph data storage provided by Neo4j is particularly suitable for the storage requirements of knowledge graphs. FlaskAPI can cooperate with the py2neo library to conveniently operate Neo4j programmatically and provide a RESTAPI interface externally. Finally, the user identity and authentication information are stored in the relational database MySQL and a high-performance interface is provided through SpringBoot. The specific process is as follows:

[0091] 1. Use Neo4j database to establish the knowledge graph data storage structure, including the definition of multiple node types and relationship types.

[0092] 2. Use the query language provided by Neo4j (such as Cypher) to perform conditional retrieval, modification and deletion operations on graph data.

[0093] 3. Develop an API interface program based on the Flask framework, load all BluePrints for Neo4j operations in the project as the controller layer, and provide py2neo's capabilities (such as conditional retrieval, modification and deletion) as REST API.

[0094] 4. Support batch import of node and relationship data from CSV files, read nodes and logical relationships in CSV template files to quickly build knowledge graphs. Support cloning graphs, copy the nodes and relationships of existing knowledge graphs, and create new course or subject graphs. Through the method of acquisition, temporary storage, and reconstruction, first obtain all nodes and relationships of the cloned target, then temporarily store them in the system memory, and finally insert the nodes and relationships in the memory to complete the cloning function.

[0095] 5. Use the SpringBoot framework to build backend services to manage and operate the MySQL database. This solves the limitation of Neo4j itself that it only has node and relationship storage. The data related to the course interface service and user identity information function is stored in the MySQL database, realizing diversified data storage and management, and providing a complete user identity and permission mechanism.

[0096] Through the implementation of the above technical solutions, the present invention successfully solves the operation and storage problems of multiple data sources in terms of knowledge graph relationship storage, operation flexibility and external interface services, and provides more comprehensive and powerful back-end support for knowledge graph applications.

[0097] In order to achieve the fourth object of the present invention, the following steps are included:

[0098] S1, obtain the course nodes and relationship content of the knowledge graph, use Echarts to render different node types accordingly, and improve the visualization effect of the knowledge graph and the recognition of the node type hierarchy;

[0099] S2, in the graph display component, register the click events of each node and relationship line, dynamically switch to the next level of nodes and relationships of the clicked element by clicking or dragging the nodes and relationship lines, and obtain the node and relationship data saved in the graph database in the backend FlaskAPI, and dynamically render the returned results to the knowledge graph display area, making the interactive experience simple and intuitive;

[0100] S3, in the graph editing component, register the click events of nodes and relationship lines to realize the click selection of nodes or relationships to be edited or deleted. After the operation, it will take effect immediately and dynamically on the graph editing component. The operation is convenient and the feedback is intuitive.

[0101] The above are only preferred specific implementation modes of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical solutions and inventive concepts of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. An intelligent construction and interactive display system for course knowledge graphs, characterized in that: The purposes include: Design the course ontology to ensure the systematic and comprehensiveness of course knowledge; This paper provides an intelligent course knowledge extraction method based on a large model, which aims to intelligently extract the key points of knowledge in the course, thereby improving the efficiency and accuracy of knowledge extraction; Provides multi-source database storage and convenient data access operations, solving the problem that traditional graph databases store single content and are inconvenient for clients to operate data, and realizing efficient storage and management of graph databases and relational databases in multi-source data scenarios; Design a knowledge graph interactive system that can be dynamically displayed and operated, which can better display the nodes and relationships stored in the knowledge graph, edit and delete the nodes and relationships, and improve the visualization effect and operation experience.

2. According to claim 1, the intelligent construction and interactive display system of course knowledge graph is characterized in that: The course ontology combines the course type, textbook catalog and internal logic, and is divided into five entities: course type, course, knowledge system, knowledge unit and knowledge key points according to different knowledge granularity; Course type: Course type refers to the type of course, which is intended to distinguish and standardize different types of courses, such as basic courses, advanced courses, and special courses. Course type helps to clarify the positioning of courses in teaching activities, and by establishing associations with specific course entities, ensures that the classification of each course is clear and the attributes are standardized; The course: a specific course entity, including the course name, course introduction, reference materials, course objectives, learning difficulty, course credits, assessment method, pilot courses and related learning resources. Course entities are linked through attributes (such as "pilot courses") and correspond to "course types", which usually correspond to specific course content in the textbook, and provide detailed information on credits, lecturers and related resources to ensure the unity of learning objectives and assessment standards; The knowledge system is used to define the level of the abstract knowledge system in the course, including the name of the knowledge system, the importance, the next knowledge system, and the attributes of the course to which it belongs. The knowledge systems are connected in order through the "next knowledge system" attribute, showing the progressive level and logical structure of knowledge. Through such division, the knowledge system clarifies the organization of the core knowledge in each course, making its content more systematic; The knowledge unit: the knowledge unit is a further refinement of the knowledge system, usually corresponding to the program directory in the textbook. Each knowledge unit entity includes the knowledge unit name, knowledge unit content, importance, learning difficulty, and the attributes of the knowledge system to which it belongs. The knowledge unit entity is hierarchically managed through the knowledge system to which it belongs; The multiple knowledge points are the smallest knowledge granularity in the course ontology. Each knowledge point represents a specific knowledge point and skill. The knowledge point entity includes the name of the knowledge point, the content of the knowledge point, the importance, the learning difficulty, the learning objectives, and the attributes of the knowledge unit to which it belongs. The knowledge point entity is associated with the parent knowledge unit through the attributes of the knowledge unit to which it belongs.

3. According to claim 1, the intelligent construction and interactive display system of course knowledge graph is characterized in that: In order to realize the hierarchical organization and automatic management of knowledge, the system introduces six key relationship types, namely, the type of course, the predecessor course, the course to which it belongs, the next knowledge system, the knowledge system to which it belongs, and the knowledge unit to which it belongs. The following describes these relationships in detail and their applications at each level; The course type is used to connect the course and its type to identify the specific category to which each course belongs; through this relationship, the system can effectively classify and manage courses; The said pre-courses: reveal the learning sequence between courses and clarify the prerequisite courses that need to be mastered before learning a certain course. Through this relationship, the system can ensure the rationality of the learning path and prevent students from entering difficult courses without the necessary basic knowledge; The courses to which it belongs are used to associate a knowledge system with its courses. Each course contains one or more knowledge systems, and a knowledge system represents the core content and knowledge framework of the course. The next knowledge system: defines the order between knowledge systems, ensuring that students can follow the correct logic or difficulty order when learning a certain knowledge system. Through this relationship, the system can plan a coherent learning path for students to avoid knowledge gaps; The knowledge system to which it belongs is used to connect the knowledge unit and the knowledge system to which it belongs, so as to identify which knowledge system the specific knowledge unit belongs to. The knowledge unit is a further subdivision of the knowledge system and covers more specific content. The belonging knowledge unit: connect the knowledge points with the corresponding knowledge units to clarify which knowledge unit the specific knowledge points belong to. As the smallest unit of the knowledge graph, the knowledge points usually represent specific concepts or skills.

4. According to claim 1, the intelligent construction and interactive display system of course knowledge graph is characterized in that: The intelligent course knowledge extraction method based on a large model is specifically divided into three steps: knowledge division and precise extraction, large model-assisted attribute identification and knowledge summary, and knowledge graph relationship construction and hierarchical organization; Knowledge division and accurate extraction of course content based on regular expressions: Based on ontology design and rule-driven, regular expressions are used to extract courses, knowledge systems, knowledge units and key points from texts, providing basic data support for building a complete knowledge graph and ensuring that the knowledge graph structure is clear and orderly; Big model assists entity attribute recognition and knowledge key point summary: In the specific implementation process of the present invention, regular expressions first realize the preliminary extraction of course content, but for complex knowledge key points and attributes, the accuracy of relying solely on rule matching is limited. Therefore, the system introduces a big model and uses natural language processing technology to further identify entity attributes and summarize knowledge key points; Knowledge graph relationship construction and hierarchical organization: Using the course data obtained in the previous step, a unified knowledge system name is set as the basic structure, and three entities, namely knowledge system, knowledge unit and knowledge key points, are structured and organized. Under the ontology design, specific attributes are preset for each entity, including sequential numbering, to achieve modularization and systematic processing of knowledge content. Finally, the course content is decomposed into standardized data files level by level, laying the foundation for the hierarchical organization and visualization construction of the knowledge graph; Through the above steps, the course content is decomposed, extracted, and organized into standardized data files, laying the foundation for the visual construction of the course knowledge graph.

5. According to claim 1, the intelligent construction and interactive display system of course knowledge graph is characterized in that: In order to achieve the third purpose of the present invention, the Neo4j graph database is used for storage, py2neo is used to perform programmatic graph data operations, and finally the FlaskAPI is used to provide an external calling interface. The graph data storage provided by Neo4j is particularly suitable for the storage needs of knowledge graphs. The FlaskAPI can cooperate with the py2neo library to conveniently operate Neo4j in a programmatic manner and provide a REST API interface to the outside. Finally, the user identity and authentication information are stored in the relational database MySQL and a high-performance interface is provided through SpringBoot.

6. According to claim 1, the intelligent construction and interactive display system of course knowledge graph is characterized in that: In order to achieve the fourth object of the present invention, the following steps are included: S1, obtain the course nodes and relationship content of the knowledge graph, use Echarts to render different node types accordingly, and improve the visualization effect of the knowledge graph and the recognition of the node type hierarchy; S2, in the graph display component, register the click events of each node and relationship line, dynamically switch to the next level of nodes and relationships of the clicked element by clicking or dragging the nodes and relationship lines, and obtain the node and relationship data saved in the graph database in the backend FlaskAPI, and dynamically render the returned results to the knowledge graph display area, making the interactive experience simple and intuitive; S3, in the graph editing component, register the click events of nodes and relationship lines to realize the click selection of nodes or relationships to be edited or deleted. After the operation, it will take effect immediately and dynamically on the graph editing component. The operation is convenient and the feedback is intuitive.

Images