Intelligent knowledge graph construction method, system and equipment for production and education fusion platform

Abstract

The invention relates to the technical field of knowledge graph construction, in particular to an intelligent knowledge graph construction method, system and equipment for a production and education fusion platform. The method comprises the following steps: firstly, acquiring educational resource data, and processing the educational resource data by adopting a preset classification rule to obtain a hierarchical knowledge point type; then identifying content features of each knowledge point, extracting associated attributes, and generating label information; then, based on the knowledge point type and the label information, the knowledge graph quality is evaluated by testing the knowledge coverage degree and the demand matching degree; after the evaluation is completed, the system automatically triggers an updating instruction, and dynamically adjusts the knowledge graph according to an optimization instruction; through the mode, the integrity of a knowledge system is ensured, organic combination of theory and practice is realized, and meanwhile, a scientific quality evaluation and optimization mechanism is established, so that the knowledge graph can be continuously optimized and perfected according to actual requirements.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of knowledge graph construction, in particular to a production-teaching integration platform intelligent knowledge graph construction method, system and device. BACKGROUND

[0002] With the in-depth development of education informatization and production-teaching integration, how to effectively manage and utilize education resources, and realize the organic combination of theoretical teaching and practical application, has become an important issue in the current education field. As an important tool for knowledge organization and management, intelligent knowledge graph plays an increasingly important role in education resource integration and knowledge transmission.

[0003] The prior art usually adopts a rule-based knowledge extraction system and a statistical-based knowledge correlation analysis method to construct a knowledge graph. These methods extract knowledge entities through pre-set rule templates, and use statistical algorithms to establish the correlation between knowledge, form a knowledge network structure, and realize the organization and management of education resources.

[0004] However, the knowledge classification of the prior art is too fixed, the knowledge correlation analysis lacks consideration of actual application scenarios, and it is difficult to adapt to dynamic changes and the education needs of production-teaching integration. This situation needs to be further improved. SUMMARY

[0005] In order to solve the problem that the existing knowledge classification is too fixed, the knowledge correlation analysis lacks consideration of actual application scenarios, and it is difficult to adapt to dynamic changes and the education needs of production-teaching integration, the present application provides a production-teaching integration platform intelligent knowledge graph construction method, system and device, which adopts the following technical solutions: In a first aspect, the present application provides a production-teaching integration platform intelligent knowledge graph construction method, comprising the following steps: Obtain education resource data, classify the education resource data according to pre-set knowledge classification rules, and obtain corresponding knowledge point types; Obtain knowledge content corresponding to the knowledge point types, identify the correlation attributes of the knowledge content, and obtain label information representing knowledge point features; According to the knowledge point types and the label information, obtain test data for testing the quality of the knowledge graph, the test data including knowledge coverage data and demand matching degree data; Obtain knowledge graph evaluation results based on the knowledge coverage data and demand matching degree data, and trigger a knowledge graph update instruction according to the evaluation results; Based on the knowledge graph update instruction, trigger a knowledge optimization instruction, and perform dynamic adjustment of the knowledge graph.

[0006] By adopting the technical scheme, in order to solve the problems of knowledge classification solidification, knowledge correlation separation from practice and imperfect quality evaluation mechanism in the traditional knowledge graph construction process; the application firstly acquires education resource data, processes by using a preset classification rule, and obtains a hierarchical knowledge point type; then identifies the content features of each knowledge point, extracts its difficulty, dependency relationship, application scenario and other correlation attributes, and generates label information; then, based on the knowledge point type and the label information, the knowledge graph quality is evaluated by testing the knowledge coverage and the demand matching degree; after the evaluation is completed, the system automatically triggers an update instruction, and dynamically adjusts the knowledge graph according to the optimization instruction; in this way, the integrity of the knowledge system is ensured, the organic combination of theory and practice is realized, and a scientific quality evaluation and optimization mechanism is established, so that the knowledge graph can be continuously optimized and improved according to the actual demand.

[0007] Optionally, the education resource data is acquired, and the education resource data is classified and processed according to a preset knowledge classification rule to obtain corresponding knowledge point types, specifically including the following steps: The historical education resource data is acquired, and knowledge point structure information is acquired from the historical education resource data; According to the knowledge point structure information, knowledge point key feature information used for knowledge graph construction is extracted from the knowledge point structure information; According to a preset knowledge classification rule, the knowledge point key feature information is classified to obtain corresponding knowledge point types.

[0008] By adopting the technical scheme, the application firstly collects and organizes historical education resource data, extracts knowledge point structure information therefrom by using data mining technology, including hierarchical relationship, correlation degree and use frequency between knowledge points; then, the structure information is analyzed in depth, and key feature information having guiding significance for knowledge graph construction is extracted, such as importance of knowledge points, application range and learning difficulty; finally, according to a preset classification rule, the knowledge points are classified scientifically in combination with the extracted key feature information.

[0009] Optionally, the knowledge point types include theoretical knowledge and practical skills; the knowledge contents corresponding to the knowledge point types are acquired, the correlation attributes of the knowledge contents are identified, and label information representing the characteristics of the knowledge points is obtained, specifically including the following steps: The knowledge contents are acquired, and the correlation attributes of the knowledge contents are extracted, wherein the correlation attributes include knowledge difficulty, knowledge dependency and application scenario; If the knowledge point type is theoretical knowledge, corresponding theoretical knowledge label information is acquired based on the correlation attributes; If the knowledge point type is practical skill, corresponding practical skill label information is acquired based on the correlation attributes; The label information includes theoretical knowledge label information and practical skill label information.

[0010] By adopting the technical solution, the knowledge content is acquired first, and the associated attributes including basic features such as knowledge difficulty, knowledge dependency and application scenario are extracted. Then, the knowledge points are processed according to the types, the concept system, the derivation process and the application conditions of the theoretical knowledge are highlighted and labeled, and the theoretical knowledge label information is generated. The operation points, tool requirements and implementation environment of the practical skills are highlighted and labeled, and the practical skill label information is generated. This classification and labeling method not only accurately reflects the features of different types of knowledge points, but also establishes a bridge between theory and practice, effectively improving the guiding value of the knowledge graph in actual teaching.

[0011] Optionally, according to the knowledge point type and the label information, test data for testing the quality of the knowledge graph is acquired, specifically including the following steps: According to the knowledge dependency and knowledge difficulty attributes in the theoretical knowledge label information and the practical skill label information, knowledge coverage data is generated; According to the application scenario attributes in the theoretical knowledge label information and the practical skill label information, demand matching degree data is generated; The knowledge coverage data includes knowledge structure integrity data and knowledge correlation integrity data; and the demand matching degree data includes theoretical knowledge matching degree data and practical skill matching degree data.

[0012] By adopting the technical solution, since the existing evaluation method often pays too much attention to the number of knowledge points or the coverage of a single dimension, ignoring the integrity of the knowledge structure and the matching degree of the actual demand, the evaluation result is difficult to truly reflect the practical value of the knowledge graph. Based on the label information of the theoretical knowledge and the practical skill, the knowledge dependency relationship and the difficulty level are analyzed, the coverage data including the knowledge structure integrity and the knowledge correlation integrity is generated, and the systematicness of the knowledge system is ensured. Then, according to the application scenario attributes in the label information, the matching degree of the theoretical knowledge and the practical skill with the actual demand is evaluated respectively, and the comprehensive demand matching degree data is generated. Through multi-dimensional data analysis, not only the structural defects in the knowledge system can be found, but also the matching degree of the knowledge content with the actual application demand can be accurately evaluated, providing a reliable basis for the optimization of the knowledge graph.

[0013] Optionally, based on the knowledge coverage data and the demand matching degree data, a knowledge graph evaluation result is acquired, and according to the evaluation result, a knowledge graph update instruction is triggered, specifically including the following steps: Based on the knowledge coverage data and the demand matching degree data, a knowledge graph evaluation instruction is triggered; Based on the knowledge graph evaluation instruction, a knowledge graph evaluation result is obtained, and the knowledge graph evaluation result includes knowledge system integrity and knowledge application matching degree. According to the knowledge graph evaluation result, a knowledge graph update instruction is triggered.

[0014] By adopting the above technical solution, the application first triggers a systematic evaluation instruction based on knowledge coverage data and demand matching data, and starts a comprehensive evaluation process. Then, through the execution of the evaluation instruction, the integrity level of the knowledge system and the matching degree of the knowledge application are comprehensively analyzed to form an evaluation result containing quantitative indicators and qualitative analysis. Finally, according to the problems and optimization directions reflected in the evaluation result, the corresponding knowledge graph update instruction is automatically triggered. By establishing the linkage mechanism among the evaluation data, the evaluation result and the update instruction, the automatic connection of knowledge graph evaluation and update is realized.

[0015] Optionally, according to the knowledge graph evaluation result, a knowledge graph update instruction is triggered, specifically including the following steps: According to the knowledge system integrity and the knowledge application matching degree, knowledge graph optimization demand information is obtained. According to the knowledge graph optimization demand information and the label information, a knowledge graph update plan is generated. Based on the knowledge graph update plan, a knowledge graph update instruction is triggered.

[0016] By adopting the above technical solution, the traditional updating method often directly modifies according to the evaluation result, lacking systematic demand analysis and plan design. The application first identifies the key fields that need to be optimized in the knowledge graph by analyzing the knowledge system integrity and the knowledge application matching degree, and forms specific optimization demand information. Then, combined with the existing label information, the association relationship and application characteristics between knowledge points are fully considered, and an update plan containing optimization target, implementation steps and expected effect is made. Finally, based on the detailed update plan, the corresponding update instruction is triggered to guide the optimization work of the knowledge graph. The updating process of the knowledge graph is more orderly and efficient.

[0017] Optionally, based on the knowledge graph update instruction, a knowledge optimization instruction is triggered to dynamically adjust the knowledge graph, specifically including the following steps: Based on the knowledge graph update instruction, the knowledge graph update progress is obtained. The knowledge graph update progress and the knowledge coverage data are input into a preset knowledge optimization model, and a preset knowledge optimization rule is obtained according to the knowledge optimization model. According to the knowledge optimization rule, a knowledge optimization instruction is triggered to realize dynamic adjustment of the knowledge graph.

[0018] By adopting the technical scheme, the application first acquires the update progress information of the knowledge graph through the monitoring system, and real-time grasps the execution of the optimization work; then inputs the update progress data and the knowledge coverage data into the preset knowledge optimization model, generates the optimization rules conforming to the characteristics of the current optimization stage through model analysis; finally, according to the dynamically generated optimization rules, triggers the corresponding optimization instructions, and realizes the real-time adjustment of the knowledge graph; makes the optimization process of the knowledge graph more flexible and accurate, and can quickly respond to new situations and new demands in the optimization process.

[0019] In a second aspect, the application provides an intelligent knowledge graph construction system of a production-teaching integration platform, comprising: A knowledge point type acquisition module is configured to acquire education resource data, perform classification processing on the education resource data according to a preset knowledge classification rule, and obtain corresponding knowledge point types, wherein the knowledge point types include theoretical knowledge and practical skills. A feature label information acquisition module is configured to acquire knowledge content corresponding to the knowledge point types, identify the associated attributes of the knowledge content, and obtain label information representing the features of the knowledge points. A test data acquisition module is configured to acquire test data for testing the quality of the knowledge graph according to the knowledge point types and the label information, wherein the test data includes knowledge coverage data and demand matching degree data. A knowledge graph update instruction triggering module is configured to acquire a knowledge graph evaluation result based on the knowledge coverage data and the demand matching degree data, and trigger a knowledge graph update instruction according to the evaluation result. A knowledge graph dynamic adjustment module is configured to trigger a knowledge optimization instruction based on the knowledge graph update instruction, and perform dynamic adjustment of the knowledge graph.

[0020] In a third aspect, the application provides an electronic device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, wherein the processor implements the steps of the intelligent knowledge graph construction method of the production-teaching integration platform when executing the computer program.

[0021] In summary, the application has at least one of the following beneficial technical effects: 1. The application first acquires education resource data, processes it using preset classification rules, and obtains hierarchical knowledge point types. Then it identifies the content characteristics of each knowledge point, extracts its difficulty, dependency relationship, application scenario, and other related attributes, and generates label information. Next, based on the knowledge point types and label information, it assesses the quality of the knowledge graph by testing the knowledge coverage and demand matching degree. After the assessment is complete, the system automatically triggers the update instruction and dynamically adjusts the knowledge graph according to the optimization instruction. In this way, the integrity of the knowledge system is guaranteed, the organic combination of theory and practice is achieved, and a scientific quality evaluation and optimization mechanism is established, enabling the knowledge graph to be continuously optimized and improved according to actual needs. 2. The application first acquires knowledge content and extracts its related attributes, including knowledge difficulty, knowledge dependency, and application scenario. Then it processes the knowledge points according to their types. For theoretical knowledge, it focuses on labeling its concept system, derivation process, and application conditions, and generates theoretical knowledge label information. For practical skills, it emphasizes labeling its operation points, tool requirements, and implementation environment, and generates practical skill label information. This classification and labeling method not only accurately reflects the characteristics of different types of knowledge points, but also establishes a bridge between theory and practice, effectively improving the guiding value of the knowledge graph in actual teaching. 3. The application first automatically triggers a systematic evaluation instruction based on knowledge coverage data and demand matching data, and starts a comprehensive evaluation process. Then it analyzes the integrity level of the knowledge system and the matching degree of knowledge application through the execution of the evaluation instruction, and forms an evaluation result containing quantitative indicators and qualitative analysis. Finally, it automatically triggers the corresponding knowledge graph update instruction according to the problems and optimization directions reflected in the evaluation result. By establishing a linkage mechanism between evaluation data, evaluation results, and update instructions, the application realizes the automatic connection of knowledge graph evaluation and update. BRIEF DESCRIPTION OF DRAWINGS

[0022] Figure 1 is a flowchart of an embodiment of the application, a production-education integration platform intelligent knowledge graph construction method; Figure 2 is a flowchart of step S100 in the production-education integration platform intelligent knowledge graph construction method of the application; Figure 3 is a flowchart of step S200 in the production-education integration platform intelligent knowledge graph construction method of the application; Figure 4 is a flowchart of step S300 in the production-education integration platform intelligent knowledge graph construction method of the application; Figure 5 is a flowchart of step S400 in the production-education integration platform intelligent knowledge graph construction method of the application; Figure 6 is a flowchart of step S430 in the method for constructing an intelligent knowledge graph of a production-education integration platform according to an embodiment of the present application; Figure 7 is a flowchart of step S500 in the method for constructing an intelligent knowledge graph of a production-education integration platform according to an embodiment of the present application; Figure 8 is a module diagram of a system for constructing an intelligent knowledge graph of a production-education integration platform according to an embodiment of the present application; Figure 9 is an internal structure diagram of an electronic device according to an embodiment of the present application. DETAILED DESCRIPTION

[0023] The terms used in the following embodiments of the present application are only for the purpose of describing specific embodiments and are not intended to be limiting of the present application. As used in the specification and the appended claims of the present application, the singular forms “a,” “an” and “the” are intended to include both singular and plural forms, unless the context clearly indicates otherwise. It will be further understood that the terms “and / or” as used herein refers to any or all possible combinations of one or more of the associated listed items.

[0024] Hereinafter, the terms “first” and “second” are only for the purpose of description and cannot be understood as implying or suggesting relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined with “first” and “second” can explicitly or implicitly include one or more of the features, and in the description of the embodiments of the present application, the meaning of “a plurality of” is two or more, unless otherwise specified.

[0025] The embodiments of the present application are further described in detail below with reference to the accompanying drawings.

[0026] In a first aspect, the present application provides a method for constructing an intelligent knowledge graph of a production-education integration platform, referring to Figure 1 , comprising the following steps: S100, obtaining education resource data, classifying the education resource data according to a preset knowledge classification rule to obtain corresponding knowledge point types.

[0027] In this embodiment, the education resource data refers to the teaching content from the college curriculum system, enterprise training materials, industry technical standards and professional certification requirements. The knowledge classification rule refers to a pre-established classification mapping table, which contains the corresponding relationship between knowledge characteristics and types. The knowledge point types include two categories of theoretical knowledge and practical skills.

[0028] Specifically, first, a knowledge feature mapping table is established, and content containing calculation formulas, principle elaboration, and concept definition is mapped to theoretical knowledge, and content containing operation steps, tool use, and experimental process is mapped to practical skills. Then, a text analysis method is used to extract keywords and semantic features of the educational resources, and the mapping table is matched to determine the knowledge point type. For example, in a hydraulic transmission course of a mechanical major, "Pascal's law and its application" is identified as theoretical knowledge, and "hydraulic system assembly process" is identified as practical skills.

[0029] In S200, knowledge content corresponding to the knowledge point type is obtained, and the associated attributes of the knowledge content are identified to obtain label information representing the characteristics of the knowledge point.

[0030] In this embodiment, the knowledge content refers to the specific teaching content after classification. The associated attributes include three dimensions of knowledge difficulty, knowledge dependency, and application scenario. The label information is divided into two categories of theoretical knowledge label information and practical skill label information.

[0031] Specifically, a knowledge attribute database is established, in which the knowledge difficulty is divided into three levels of basic, advanced, and advanced, the knowledge dependency is represented by a pre-knowledge relationship graph, and the application scenario is defined by a scene description word library. For theoretical knowledge, its concept system, derivation process, and application conditions are extracted to generate labels; for practical skills, its operation points, tool requirements, and implementation environment are extracted to generate labels. For example, in a numerical control machining course, a knowledge dependency chain including "machining precision requirement-tool selection-machining process" is established, and the application scenario of "workshop training" is labeled.

[0032] In S300, test data for testing the quality of the knowledge graph are obtained according to the knowledge point type and the label information, and the test data include knowledge coverage data and demand matching degree data.

[0033] In this embodiment, the knowledge coverage data includes knowledge structure integrity data and knowledge association integrity data. The demand matching degree data includes theoretical knowledge matching degree data and practical skill matching degree data. The test data are used to evaluate the quality of the knowledge graph.

[0034] Specifically, a knowledge point evaluation index library is established, the coverage data is generated by calculating the knowledge point coverage rate and the knowledge point correlation degree, and the matching degree data is generated by analyzing the teaching target achievement degree and the post skill coincidence degree. The scoring matrix is used to quantify each index, and the threshold range is set to determine the test result.

[0035] In S400, a knowledge graph evaluation result is obtained based on the knowledge coverage data and the demand matching degree data, and according to the evaluation result, a knowledge graph update instruction is triggered.

[0036] In this embodiment, the knowledge graph evaluation result includes two dimensions of knowledge system integrity and knowledge application matching degree. The knowledge graph update instruction is used to trigger the graph optimization process.

[0037] S500, based on the knowledge graph update instruction, trigger the knowledge optimization instruction, and perform dynamic adjustment of the knowledge graph.

[0038] In this embodiment, the knowledge optimization instruction includes an optimization target, an optimization range, and an optimization method. The dynamic adjustment of the knowledge graph refers to real-time updating of knowledge content and structure.

[0039] Specifically, an optimization rule library is established, a corresponding optimization strategy is selected according to the type of the update instruction, optimization parameters are determined through data analysis, and dynamic adjustment of the knowledge graph is realized. In the optimization process, an incremental updating method is used to maintain the stability of the original knowledge structure.

[0040] In one embodiment, referring to Figure 2 , in step S100, the educational resource data is obtained, the educational resource data is classified according to the preset knowledge classification rule, and the corresponding knowledge point type is obtained, which specifically includes the following steps: S110, obtaining historical educational resource data, and obtaining knowledge point structure information according to the historical educational resource data.

[0041] In this embodiment, the historical educational resource data refers to the structured teaching content in the curriculum outline, teaching material content, training manual, and teaching courseware in the past three years. The knowledge point structure information includes information about the hierarchical relationship of knowledge points, the composition elements of knowledge points, and the connection relationship of knowledge points.

[0042] S120, according to the knowledge point structure information, extracting knowledge point key feature information used for knowledge graph construction from the knowledge point structure information.

[0043] In this embodiment, the knowledge point key feature information includes two types of information: knowledge content features and knowledge application features. The knowledge content features describe the core content and basic attributes of the knowledge points, and the knowledge application features describe the practical requirements and application scenarios of the knowledge points.

[0044] Specifically, a feature extraction rule library is established, which includes three types of rules: keyword matching rules, semantic analysis rules, and context association rules. The knowledge point structure information is filtered through the rule library to extract key features.

[0045] S130, according to the preset knowledge classification rule, classifying the knowledge point key feature information to obtain the corresponding knowledge point type.

[0046] In the embodiment, the preset knowledge classification rule refers to a classification standard system based on knowledge characteristics. The knowledge point type refers to a knowledge point attribute determined according to the classification rule.

[0047] Specifically, a knowledge classification decision tree is constructed, and knowledge point key feature information is matched with a judgment node of the decision tree. The first layer of the decision tree is based on knowledge attribute judgment theory or skill, and the second layer is based on knowledge depth judgment basis or profession.

[0048] In one embodiment, with reference to Figure 3 , the knowledge point type includes theoretical knowledge and practical skill; in step S200, knowledge content corresponding to the knowledge point type is acquired, an associated attribute of the knowledge content is identified, and label information representing a knowledge point feature is obtained, specifically including the following steps: S210, acquiring knowledge content and extracting an associated attribute of the knowledge content.

[0049] The associated attribute includes knowledge difficulty, knowledge dependency, and application scenario.

[0050] In the embodiment, the knowledge content refers to specific teaching content of each knowledge point, including three parts of definition and explanation, core points, and application methods of the knowledge point. In the associated attribute, the knowledge difficulty is divided into three levels of primary, intermediate, and advanced; the knowledge dependency refers to prerequisite knowledge that needs to be mastered for the knowledge point; and the application scenario refers to an actual application environment of the knowledge point.

[0051] Specifically, an associated attribute identification library is established, including a difficulty evaluation standard table, a knowledge dependency relationship table, and an application scenario description table. The difficulty evaluation standard table divides the knowledge point into three levels of “understanding-application-innovation” based on cognitive levels; the knowledge dependency relationship table records the prerequisite relationship between knowledge points; and the application scenario description table lists three types of scene characteristics of “classroom teaching, laboratory operation, and enterprise practice”. Key information is extracted from the knowledge content by a text analysis tool, matched with the identification library, and each associated attribute is determined.

[0052] S220, if the knowledge point type is theoretical knowledge, corresponding theoretical knowledge label information is acquired based on the associated attribute.

[0053] In the embodiment, the theoretical knowledge label information includes three types of concept label, principle label, and method label. The concept label describes the basic definition, the principle label describes the internal law, and the method label describes the theoretical application mode.

[0054] Specifically, a theoretical knowledge label generation rule set is constructed, which includes a label extraction rule and a label combination rule. The label extraction rule identifies key concept words, principle description words, and method description words based on text features; and the label combination rule defines a combination mode between multiple labels. The associated attribute information is processed through the rule set to generate a label combination conforming to the characteristics of theoretical knowledge.

[0055] S230, if the knowledge point type is practical skill, the corresponding practical skill label information is obtained based on the association attribute.

[0056] In this embodiment, the practical skill label information includes three types of operation label, tool label and environment label. The operation label describes the specific operation steps, the tool label describes the required equipment tools, and the environment label describes the operation environment requirements.

[0057] Specifically, a practical skill label mapping table is established, which includes an operation element mapping set, a tool requirement mapping set, and an environment condition mapping set. The association attribute is converted into standardized skill labels through the mapping table.

[0058] In one embodiment, referring to Figure 4 , in step S300, test data for testing the quality of the knowledge graph is obtained according to the knowledge point type and the label information, specifically including the following steps: S310, knowledge coverage data is generated according to the knowledge dependency and knowledge difficulty attributes in the theoretical knowledge label information and the practical skill label information.

[0059] Specifically, a knowledge coverage calculation matrix is established, which includes a knowledge point distribution table and a knowledge association table. The knowledge point distribution table records the number distribution of knowledge points of different difficulty levels; the knowledge association table records the dependency relationship strength between knowledge points. The coverage data is generated by calculating the coverage rate index in the matrix, which includes knowledge point coverage rate, difficulty level coverage rate, and knowledge dependency coverage rate.

[0060] S320, requirement matching degree data is generated according to the application scenario attribute in the theoretical knowledge label information and the practical skill label information.

[0061] Specifically, a requirement matching degree evaluation table is constructed, which includes a scene feature list and a requirement corresponding rule. The scene feature list lists the key features of different application scenarios; the requirement corresponding rule defines the matching standard of knowledge points and application requirements. The matching degree data is generated by feature matching calculation, which considers three factors of scene applicability, application frequency and implementation condition.

[0062] S330, the knowledge coverage data includes knowledge structure integrity data and knowledge association integrity data; the requirement matching degree data includes theoretical knowledge matching degree data and practical skill matching degree data.

[0063] Specifically, a test data analysis framework is established, which includes an integrity evaluation module and a matching degree evaluation module. The integrity evaluation module generates integrity data by calculating the proportion of knowledge points, the proportion of hierarchical distribution, and the correlation strength coefficient. The matching degree evaluation module generates matching degree data by analyzing the theoretical content applicability and skill requirement compliance.

[0064] In one embodiment, with reference to Figure 5 In step S400, knowledge graph evaluation results are obtained based on the knowledge coverage data and the demand matching degree data. According to the evaluation results, a knowledge graph update instruction is triggered, which specifically includes the following steps: S410, based on the knowledge coverage data and the demand matching degree data, a knowledge graph evaluation instruction is triggered.

[0065] In this embodiment, the knowledge graph evaluation instruction refers to a system command for starting the evaluation process. The triggering conditions of the evaluation instruction include a periodic triggering condition and a threshold triggering condition. The periodic triggering condition is set to once every semester, and the threshold triggering condition is set to automatically trigger when the coverage degree is less than 85% or the matching degree is less than 80%.

[0066] Specifically, an evaluation trigger mechanism table is established, which includes time trigger rules and data trigger rules. The time trigger rules define the fixed evaluation period, and the data trigger rules set the warning thresholds of coverage and matching degree. The triggering conditions are checked periodically by the monitoring system, and the evaluation instruction is automatically generated when the triggering conditions are met. For example, after the update of a certain professional curriculum system, the system detects that the knowledge coverage degree has decreased to 83%, which is lower than the warning threshold, and automatically triggers the evaluation instruction for comprehensive evaluation.

[0067] S420, based on the knowledge graph evaluation instruction, the knowledge graph evaluation results are obtained, which include the knowledge system integrity and the knowledge application matching degree.

[0068] In this embodiment, the knowledge system integrity refers to the systematization and completeness score of the knowledge structure. The knowledge application matching degree refers to the degree of fit between the knowledge content and the actual application demand.

[0069] Specifically, an evaluation score standard table is constructed, which includes integrity scoring rules and matching degree scoring rules. The integrity scoring rules are based on the knowledge point coverage, the knowledge level distribution, and the knowledge correlation strength for scoring. The matching degree scoring rules are based on the theoretical applicability, the practical feasibility, and the demand satisfaction for scoring. The scores of each dimension are integrated by a weighted calculation method to generate the final evaluation results.

[0070] S430, according to the knowledge graph evaluation results, a knowledge graph update instruction is triggered.

[0071] Specifically, an update instruction generation rule library is established, which includes evaluation result judgment criteria and an update strategy mapping table. The judgment criteria divides the evaluation scores into different intervals, corresponding to different update requirements; the update strategy mapping table determines the update method according to the specific performance of the evaluation result. A specific update instruction is generated through rule matching.

[0072] In one embodiment, with reference to Figure 6 In step S430, according to the knowledge graph evaluation result, a knowledge graph update instruction is triggered, specifically including the following steps: S431, obtaining knowledge graph optimization demand information according to the knowledge system completeness and the knowledge application matching degree.

[0073] In this embodiment, the knowledge graph optimization demand information includes two parts: optimization target and optimization content. The optimization target includes completeness improvement target and matching degree improvement target; the optimization content includes three aspects: knowledge points that need to be supplemented, knowledge structure that needs to be adjusted, and application scenarios that need to be updated.

[0074] S432, generating a knowledge graph update plan according to the knowledge graph optimization demand information and the label information.

[0075] In this embodiment, the knowledge graph update plan includes three parts: update content list, update priority order, and update execution configuration. The update content list lists the specific update items to be executed; the update priority order determines the execution order of the update items; the update execution configuration includes three elements: update method, update time node, and required resources.

[0076] Specifically, an update plan generation matrix is established, which includes priority judgment rules and execution configuration rules. In the priority judgment rules, the update content is divided into basic support type knowledge and substantive content type knowledge: the basic support type knowledge includes knowledge framework structure, knowledge correlation relationship, and knowledge classification system, and the update priority of this type of knowledge is the highest to ensure the completeness of the basic framework of the knowledge graph; the substantive content type knowledge includes specific knowledge point content, application cases, and practice requirements, and this type of knowledge is updated after the basic framework is stable. For example, when updating the artificial intelligence professional knowledge graph, first update the basic support knowledge such as “deep learning theory framework” to establish the algorithm system framework, and then update the substantive content knowledge such as “convolutional neural network application examples”. The specific execution configuration rules define the update method of different types of knowledge: the basic support type knowledge adopts full update method, and the substantive content type knowledge adopts incremental update method. Through matrix operation, a complete update plan including update order and specific configuration is generated to realize the ordered update of the knowledge graph.

[0077] S433, triggering a knowledge graph update instruction based on the knowledge graph update plan.

[0078] Specifically, based on the knowledge graph update plan, executable update instructions are generated.

[0079] In one embodiment, with reference to Figure 7 , in step S500, based on the knowledge graph update instructions, knowledge optimization instructions are triggered to perform dynamic adjustment of the knowledge graph, specifically including the following steps: S510, based on the knowledge graph update instructions, the knowledge graph update progress is obtained.

[0080] In this embodiment, the knowledge graph update progress includes three dimensions of update completion degree and update timeliness degree. The update completion degree represents the execution ratio of the update task; the update timeliness degree represents the timeliness of the update execution.

[0081] S520, the knowledge graph update progress and the knowledge coverage data are input into a preset knowledge optimization model, and the preset knowledge optimization rules are obtained according to the knowledge optimization model.

[0082] In this embodiment, the knowledge optimization model refers to an analysis model for judging the update progress and generating optimization rules. The preset knowledge optimization rules include first optimization rules and second optimization rules, which are used to guide the optimization process of the knowledge graph in different situations.

[0083] Specifically, the obtained knowledge graph update progress data and knowledge coverage data are input into the preset knowledge optimization model. In this optimization model, first, the update progress is compared with the preset progress benchmark value. When the update progress is lower than the benchmark value, it is determined which optimization rule is currently used. If the first optimization rule (e.g., incremental update rule) is being used and the optimization effect is not ideal, the second optimization rule (e.g., reconstruction update rule) is switched to. The purpose of changing the optimization rule is to avoid the situation that the current optimization rule cannot effectively improve the quality of the knowledge graph, and to improve the optimization effect by switching the optimization strategy. For example, in the optimization process of the computer network course knowledge graph, it is found that the knowledge coverage degree has not reached the expected value after using the incremental update method (first optimization rule). At this time, the model will automatically switch to the reconstruction update method (second optimization rule) to comprehensively optimize the knowledge system of the course. For example, when using the incremental update rule (first optimization rule) to process the knowledge graph, the original knowledge structure has not adapted to the new teaching needs, and only adding new knowledge points may lead to confusion in the association between knowledge. Or use the method of optimizing by chapter to update the engineering drawing course, but in actual teaching, it is necessary to break the traditional chapter structure and reorganize the knowledge points according to the engineering project practice. In this case, continuing to use the first optimization rule will lead to a large amount of resources being invested but not achieving the expected effect.

[0084] More specifically, an optimization rule switching matrix is established, which includes a progress evaluation threshold table and a rule conversion table. The progress evaluation threshold table sets the benchmark values of key indicators, such as "the update completion degree is not less than 85%, and the coverage degree is improved by not less than 10%"; the rule conversion table defines the switching conditions and conversion methods between different optimization rules. When it is found that a certain professional module fails to achieve the expected effect by using the first optimization rule (such as "optimization by chapters"), the system will automatically switch to the second optimization rule (such as "optimization by knowledge cluster reorganization") and re-attempt optimization. This dynamic switching mechanism ensures the adaptability and effectiveness of the knowledge graph optimization process.

[0085] S530, according to the knowledge optimization rule, triggering the knowledge optimization instruction, realizing the dynamic adjustment of the knowledge graph.

[0086] Specifically, according to the first optimization rule or the second optimization rule, the knowledge optimization instruction is triggered to perform incremental update or reconfiguration update on the knowledge graph, thereby realizing the dynamic adjustment of the knowledge graph.

[0087] It should be understood that the size of the serial number of each step in the above embodiment does not mean the order of execution, and the execution order of each process should be determined according to its function and inherent logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0088] In a second aspect, the present application provides an intelligent knowledge graph construction system of an industry-education integration platform. The intelligent knowledge graph construction system of the industry-education integration platform is described below in combination with the intelligent knowledge graph construction method of the industry-education integration platform.

[0089] Reference Figure 8 An intelligent knowledge graph construction system of an industry-education integration platform, comprising: A knowledge point type acquisition module configured to acquire education resource data, perform classification processing on the education resource data according to a preset knowledge classification rule, and obtain corresponding knowledge point types, wherein the knowledge point types include theoretical knowledge and practical skills; A feature label information acquisition module configured to acquire knowledge content corresponding to the knowledge point types, identify associated attributes of the knowledge content, and obtain label information representing features of the knowledge points; A test data acquisition module configured to acquire test data for testing the quality of the knowledge graph according to the knowledge point types and the label information, wherein the test data includes knowledge coverage degree data and demand matching degree data; A knowledge graph update instruction triggering module configured to acquire a knowledge graph evaluation result based on the knowledge coverage degree data and the demand matching degree data, and trigger a knowledge graph update instruction according to the evaluation result; A knowledge graph dynamic adjustment module configured to trigger a knowledge optimization instruction based on the knowledge graph update instruction, and perform dynamic adjustment of the knowledge graph.

[0090] In one embodiment, the present application provides an electronic device, which can be a server, and an internal structure diagram thereof can be as shown in Figure 9 The electronic device includes a processor, a memory and a network interface connected through a system bus. The processor of the electronic device is configured to provide computing and control capabilities. The memory of the electronic device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for running the operating system and the computer program in the non-volatile storage medium. The database of the electronic device is configured to store data. The network interface of the electronic device is configured to communicate with an external terminal through a network connection. The computer program is executed by the processor to implement a method for constructing an intelligent knowledge graph of a production-teaching integration platform.

[0091] Those skilled in the art can understand that Figure 9 The structure shown in the above embodiment is only a block diagram of part of the structure related to the scheme of the present application, and does not constitute a limitation on the electronic device to which the scheme of the present application is applied. Specifically, the electronic device can include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0092] In one embodiment, an electronic device is also provided, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the steps in the above method embodiments.

[0093] Those skilled in the art can understand that all or part of the processes in the above method embodiments can be completed by a computer program instructing related hardware. The above computer program can be stored in a non-volatile computer readable storage medium, and when executed, can include the processes of the above method embodiments. Any reference to a memory, storage, database or other medium used in the embodiments provided by the present application can include at least one of a non-volatile and volatile memory. The non-volatile memory can include a read-only memory (ROM), a magnetic tape, a floppy disk, a flash memory or an optical storage, etc. The volatile memory can include a random access memory (RAM) or an external cache memory. As an illustration but not as a limitation, the RAM can be in various forms, such as a static random access memory (SRAM) or a dynamic random access memory (DRAM), etc.

[0094] The above are all preferred embodiments of the present application, and do not limit the protection scope of the present application, so that: all equivalent changes made according to the structure, shape, principle of the present application should be covered in the protection scope of the present application.

Claims

1. A method for constructing an intelligent knowledge graph for an industry-education integration platform, characterized in that: The steps include: Acquire educational resource data, classify the educational resource data according to preset knowledge classification rules, and obtain corresponding knowledge point types; Acquire the knowledge content corresponding to the knowledge point type, identify the associated attributes of the knowledge content, and obtain label information representing the characteristics of the knowledge point; According to the knowledge point type and the label information, obtaining test data for testing the quality of the knowledge graph, the test data including knowledge coverage data and demand matching data; Obtaining a knowledge graph evaluation result based on the knowledge coverage data and the demand matching data, and triggering a knowledge graph update instruction according to the evaluation result; Based on the knowledge graph update instruction, the knowledge optimization instruction is triggered to perform dynamic adjustment of the knowledge graph.

2. The method for constructing an intelligent knowledge graph for an industry-education integration platform according to claim 1 is characterized in that: Obtaining educational resource data, classifying the educational resource data according to preset knowledge classification rules, and obtaining corresponding knowledge point types, specifically includes the following steps: Acquire historical education resource data, and acquire knowledge point structure information based on the historical education resource data; Extracting key feature information of the knowledge points for knowledge graph construction from the knowledge point structure information according to the knowledge point structure information; According to the preset knowledge classification rules, the key feature information of the knowledge point is classified to obtain the corresponding knowledge point type.

3. The method for constructing an intelligent knowledge graph for an industry-education integration platform according to claim 1 is characterized in that: The knowledge point types include theoretical knowledge and practical skills. Acquiring the knowledge content corresponding to the knowledge point type, identifying the associated attributes of the knowledge content, and obtaining label information representing the characteristics of the knowledge point specifically includes the following steps: Acquire knowledge content and extract associated attributes of the knowledge content, wherein the associated attributes include knowledge difficulty, knowledge dependency, and application scenario; If the knowledge point type is theoretical knowledge, obtaining corresponding theoretical knowledge label information based on the associated attribute; If the knowledge point type is practical skills, obtaining corresponding practical skill label information based on the associated attributes; The label information includes theoretical knowledge label information and practical skills label information.

4. The method for constructing an intelligent knowledge graph for an industry-education integration platform according to claim 3 is characterized in that: According to the knowledge point type and the label information, test data for testing the quality of the knowledge graph is obtained, specifically including the following steps: Generate knowledge coverage data based on the knowledge dependency and knowledge difficulty attributes in the theoretical knowledge label information and the practical skill label information; Generate demand matching data based on the application scenario attributes in the theoretical knowledge label information and the practical skill label information; The knowledge coverage data includes knowledge structure integrity data and knowledge association integrity data; The demand matching data includes theoretical knowledge matching data and practical skills matching data.

5. The method for constructing an intelligent knowledge graph for an industry-education integration platform according to claim 1 is characterized in that: Obtaining a knowledge graph evaluation result based on the knowledge coverage data and the demand matching data, and triggering a knowledge graph update instruction according to the evaluation result, specifically comprising the following steps: Based on the knowledge coverage data and the demand matching data, trigger a knowledge graph evaluation instruction; Based on the knowledge graph evaluation instruction, obtaining a knowledge graph evaluation result, wherein the knowledge graph evaluation result includes the completeness of the knowledge system and the matching degree of knowledge application; According to the knowledge graph evaluation result, a knowledge graph update instruction is triggered.

6. The method for constructing an intelligent knowledge graph for an industry-education integration platform according to claim 5 is characterized in that: According to the knowledge graph evaluation result, a knowledge graph update instruction is triggered, which specifically includes the following steps: Obtaining knowledge graph optimization demand information based on the completeness of the knowledge system and the matching degree of knowledge application; Generate a knowledge graph update plan based on the knowledge graph optimization demand information and the tag information; Based on the knowledge graph update plan, a knowledge graph update instruction is triggered.

7. The method for constructing an intelligent knowledge graph for an industry-education integration platform according to claim 1, characterized in that: Based on the knowledge graph update instruction, the knowledge optimization instruction is triggered to perform dynamic adjustment of the knowledge graph, which specifically includes the following steps: Based on the knowledge graph update instruction, obtain the knowledge graph update progress; Inputting the knowledge graph update progress and the knowledge coverage data into a preset knowledge optimization model, and obtaining preset knowledge optimization rules according to the knowledge optimization model; According to the knowledge optimization rules, the knowledge optimization instructions are triggered to achieve dynamic adjustment of the knowledge graph.

8. An intelligent knowledge graph construction system for an industry-education integration platform, characterized by: include: A knowledge point type acquisition module is used to acquire educational resource data, classify the educational resource data according to preset knowledge classification rules, and obtain corresponding knowledge point types, wherein the knowledge point types include theoretical knowledge and practical skills; A feature label information acquisition module is used to acquire the knowledge content corresponding to the knowledge point type, identify the associated attributes of the knowledge content, and obtain label information representing the features of the knowledge point; A test data acquisition module is used to acquire test data for testing the quality of the knowledge graph according to the knowledge point type and the label information, wherein the test data includes knowledge coverage data and requirement matching data; A knowledge graph update instruction triggering module is used to obtain a knowledge graph evaluation result based on the knowledge coverage data and the demand matching data, and trigger a knowledge graph update instruction according to the evaluation result; The knowledge graph dynamic adjustment module is used to trigger the knowledge optimization instruction based on the knowledge graph update instruction and perform dynamic adjustment of the knowledge graph.

9. An electronic device, characterized in that: It includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of the method for constructing an intelligent knowledge graph for an industry-education integration platform described in any one of claims 1 to 7 are implemented.

Images